Search

The sky may seem big enough for two flying vehicles to collide, but the facts show that mid-air collisions still occur occasionally and are a significant concern. Pilots learn manual tactics to avoid collisions, such as see-and-avoid, but these rules have limitations. Automated solutions have reduced collisions, but these technologies are not mandatory in all countries or airspaces, and they are expensive. These problems have prompted researchers to continue the search for low-cost solutions. One attractive solution is to use computer vision to detect obstacles in the air due to its reduced cost and weight. A well-trained deep learning solution is appealing because object detection is fast in most cases, but it relies entirely on the training data set. The algorithm chosen for this study is optical flow. The optical flow vectors can help us to separate the motion caused by camera motion from the motion caused by incoming objects without relying on training data. This paper describes the development of an optical flow-based airborne obstacle detection algorithm to avoid mid-air collisions. The approach uses the visual information from a monocular camera and detects the obstacles using morphological filters, optical flow, focus of expansion, and a data clustering algorithm. The proposal was evaluated using realistic vision data obtained with a self-developed simulator. The simulator provides different environments, trajectories, and altitudes of flying objects. The results showed that the optical flow-based algorithm detected all incoming obstacles along their trajectories in the experiments. The results showed an F-score greater than 75% and a good balance between precision and recall. [ABSTRACT FROM AUTHOR]

Structural engineers are often required to draw two-dimensional engineering sketches for quick structural analysis, either by hand calculation or using analysis software. However, calculation by hand is slow and error-prone, and the manual conversion of a hand-drawn sketch into a virtual model is tedious and time-consuming. This paper presents a complete and autonomous framework for converting a hand-drawn engineering sketch into an analyzed structural model using a camera and computer vision. In this framework, a computer vision object detection stage initially extracts information about the raw features in the image of the beam diagram. Next, a computer vision number-reading model transcribes any handwritten numerals appearing in the image. Then, feature association models are applied to characterize the relationships among the detected features in order to build a comprehensive structural model. Finally, the structural model generated is analyzed using OpenSees. In the system presented, the object detection model achieves a mean average precision of 99.1%, the number-reading model achieves an accuracy of 99.0%, and the models in the feature association stage achieve accuracies ranging from 95.1% to 99.5%. Overall, the tool analyzes 45.0% of images entirely correctly and the remaining 55.0% of images partially correctly. The proposed framework holds promise for other types of structural sketches, such as trusses and frames. Moreover, it can be a valuable tool for structural engineers that is capable of improving the efficiency, safety, and sustainability of future construction projects. [ABSTRACT FROM AUTHOR]

Due to the sharp increase in household waste, its separate collection is essential in order to reduce the huge amount of household waste, since it is difficult to recycle trash without separate collection. However, since it is costly and time-consuming to separate trash manually, it is crucial to develop an automatic system for separate collection using deep learning and computer vision. In this paper, we propose two Anchor-free-based Recyclable Trash Detection Networks (ARTD-Net) which can recognize overlapped multiple wastes of different types efficiently by using edgeless modules: ARTD-Net1 and ARTD-Net2. The former is an anchor-free based one-stage deep learning model which consists of three modules: centralized feature extraction, multiscale feature extraction and prediction. The centralized feature extraction module in backbone architecture focuses on extracting features around the center of the input image to improve detection accuracy. The multiscale feature extraction module provides feature maps of different scales through bottom-up and top-down pathways. The prediction module improves classification accuracy of multiple objects based on edge weights adjustments for each instance. The latter is an anchor-free based multi-stage deep learning model which can efficiently finds each of waste regions by additionally exploiting region proposal network and RoIAlign. It sequentially performs classification and regression to improve accuracy. Therefore, ARTD-Net2 is more accurate than ARTD-Net1, while ARTD-Net1 is faster than ARTD-Net2. We shall show that our proposed ARTD-Net1 and ARTD-Net2 methods achieve competitive performance in mean average precision and F1 score compared to other deep learning models. The existing datasets have several problems that do not deal with the important class of wastes produced commonly in the real world, and they also do not consider the complex arrangement of multiple wastes with different types. Moreover, most of the existing datasets have an insufficient number of images with low resolution. We shall present a new recyclables dataset which is composed of a large number of high-resolution waste images with additional essential classes. We shall show that waste detection performance is improved by providing various images with the complex arrangement of overlapped multiple wastes with different types. [ABSTRACT FROM AUTHOR]

Deep Learning (DL) has provided a significant breakthrough in many areas of research and industry. The development of Convolutional Neural Networks (CNNs) has enabled the improvement of computer vision-based techniques, making the information gathered from cameras more useful. For this reason, recently, studies have been carried out on the use of image-based DL in some areas of people's daily life. In this paper, an object detection-based algorithm is proposed to modify and improve the user experience in relation to the use of cooking appliances. The algorithm can sense common kitchen objects and identify interesting situations for users. Some of these situations are the detection of utensils on lit hobs, recognition of boiling, smoking and oil in kitchenware, and determination of good cookware size adjustment, among others. In addition, the authors have achieved sensor fusion by using a cooker hob with Bluetooth connectivity, so it is possible to automatically interact with it via an external device such as a computer or a mobile phone. Our main contribution focuses on supporting people when they are cooking, controlling heaters, or alerting them with different types of alarms. To the best of our knowledge, this is the first time a YOLO algorithm has been used to control the cooktop by means of visual sensorization. Moreover, this research paper provides a comparison of the detection performance among different YOLO networks. Additionally, a dataset of more than 7500 images has been generated and multiple data augmentation techniques have been compared. The results show that YOLOv5s can successfully detect common kitchen objects with high accuracy and fast speed, and it can be employed for realistic cooking environment applications. Finally, multiple examples of the identification of interesting situations and how we act on the cooktop are presented. [ABSTRACT FROM AUTHOR]

Object detection based on deep learning is one of the most important and fundamental tasks of computer vision. High-performance detection algorithms have been widely used in many practical fields. For the management of workers wearing helmets in construction scenarios, this paper proposes a framework model based on the YOLOv5 detection algorithm, combined with multi-object tracking algorithms, to monitor and track whether workers wear safety helmets in real-time video. The improved StrongSORT tracking algorithm of DeepSORT is selected to reduce the loss of the tracked object caused by the occlusion, trajectory blur, and motion scale of the object. The safety helmet dataset is trained with YOLOv5s, and the best result of training is used as the weight model in the StrongSORT tracking algorithm. The experimental results show that the mAP@0.5 of all classes in the YOLOv5s model can reach 95.1% in the validation dataset, mAP@0.5:0.95 is 62.1%, and the precision of wearing helmet is 95.7%. After the box regression loss function was changed from CIOU to Focal-EIOU, the mAP@0.5 increased to 95.4%, mAP@0.5:0.95 increased to 62.9%, and the precision of wearing helmet increased to 96.5%, which were increased by 0.3%, 0.8% and 0.8%, respectively. StrongSORT can update object trajectories in video frames at a speed of 0.05 s per frame. Based on the improved YOLOv5s combined with the StrongSORT tracking algorithm, the helmet-wearing tracking detection can achieve better performance. [ABSTRACT FROM AUTHOR]

Traffic sign detection is a crucial task in computer vision, finding wide-ranging applications in intelligent transportation systems, autonomous driving, and traffic safety. However, due to the complexity and variability of traffic environments and the small size of traffic signs, detecting small traffic signs in real-world scenes remains a challenging problem. In order to improve the recognition of road traffic signs, this paper proposes a small object detection algorithm for traffic signs based on the improved YOLOv7. First, the small target detection layer in the neck region was added to augment the detection capability for small traffic sign targets. Simultaneously, the integration of self-attention and convolutional mix modules (ACmix) was applied to the newly added small target detection layer, enabling the capture of additional feature information through the convolutional and self-attention channels within ACmix. Furthermore, the feature extraction capability of the convolution modules was enhanced by replacing the regular convolution modules in the neck layer with omni-dimensional dynamic convolution (ODConv). To further enhance the accuracy of small target detection, the normalized Gaussian Wasserstein distance (NWD) metric was introduced to mitigate the sensitivity to minor positional deviations of small objects. The experimental results on the challenging public dataset TT100K demonstrate that the SANO-YOLOv7 algorithm achieved an 88.7% mAP@0.5, outperforming the baseline model YOLOv7 by 5.3%. [ABSTRACT FROM AUTHOR]

Object detection and tracking in camera images is a fundamental technology for computer vision and is used in various applications. In particular, object tracking using high-speed cameras is expected to be applied to real-time control in robotics. Therefore, it is required to increase tracking speed and detection accuracy. Currently, however, it is difficult to achieve both of those things simultaneously. In this paper, we propose a tracking method that combines multiple methods: correlation filter-based object tracking, deep learning-based object detection, and motion detection with background subtraction. The algorithms work in parallel and assist each other's processing to improve the overall performance of the system. We named it the "Mutual Assist tracker of feature Filters and Detectors (MAFiD method)". This method aims to achieve both high-speed tracking of moving objects and high detection accuracy. Experiments were conducted to verify the detection performance and processing speed by tracking a transparent capsule moving at high speed. The results show that the tracking speed was 618 frames per second (FPS) and the accuracy was 86% for Intersection over Union (IoU). The detection latency was 3.48 ms. These experimental scores are higher than those of conventional methods, indicating that the MAFiD method achieved fast object tracking while maintaining high detection performance. This proposal will contribute to the improvement of object-tracking technology. [ABSTRACT FROM AUTHOR]

Visually impaired people seek social integration, yet their mobility is restricted. They need a personal navigation system that can provide privacy and increase their confidence for better life quality. In this paper, based on deep learning and neural architecture search (NAS), we propose an intelligent navigation assistance system for visually impaired people. The deep learning model has achieved significant success through well-designed architecture. Subsequently, NAS has proved to be a promising technique for automatically searching for the optimal architecture and reducing human efforts for architecture design. However, this new technique requires extensive computation, limiting its wide use. Due to its high computation requirement, NAS has been less investigated for computer vision tasks, especially object detection. Therefore, we propose a fast NAS to search for an object detection framework by considering efficiency. The NAS will be used to explore the feature pyramid network and the prediction stage for an anchor-free object detection model. The proposed NAS is based on a tailored reinforcement learning technique. The searched model was evaluated on a combination of the Coco dataset and the Indoor Object Detection and Recognition (IODR) dataset. The resulting model outperformed the original model by 2.6% in average precision (AP) with acceptable computation complexity. The achieved results proved the efficiency of the proposed NAS for custom object detection. [ABSTRACT FROM AUTHOR]

The counting of surgical instruments is an important task to ensure surgical safety and patient health. However, due to the uncertainty of manual operations, there is a risk of missing or miscounting instruments. Applying computer vision technology to the instrument counting process can not only improve efficiency, but also reduce medical disputes and promote the development of medical informatization. However, during the counting process, surgical instruments may be densely arranged or obstruct each other, and they may be affected by different lighting environments, all of which can affect the accuracy of instrument recognition. In addition, similar instruments may have only minor differences in appearance and shape, which increases the difficulty of identification. To address these issues, this paper improves the YOLOv7x object detection algorithm and applies it to the surgical instrument detection task. First, the RepLK Block module is introduced into the YOLOv7x backbone network, which can increase the effective receptive field and guide the network to learn more shape features. Second, the ODConv structure is introduced into the neck module of the network, which can significantly enhance the feature extraction ability of the basic convolution operation of the CNN and capture more rich contextual information. At the same time, we created the OSI26 data set, which contains 452 images and 26 surgical instruments, for model training and evaluation. The experimental results show that our improved algorithm exhibits higher accuracy and robustness in surgical instrument detection tasks, with F1, AP, AP50, and AP75 reaching 94.7%, 91.5%, 99.1%, and 98.2%, respectively, which are 4.6%, 3.1%, 3.6%, and 3.9% higher than the baseline. Compared to other mainstream object detection algorithms, our method has significant advantages. These results demonstrate that our method can more accurately identify surgical instruments, thereby improving surgical safety and patient health. [ABSTRACT FROM AUTHOR]

Industrial inspection is crucial for maintaining quality and safety in industrial processes. Deep learning models have recently demonstrated promising results in such tasks. This paper proposes YOLOX-Ray, an efficient new deep learning architecture tailored for industrial inspection. YOLOX-Ray is based on the You Only Look Once (YOLO) object detection algorithms and integrates the SimAM attention mechanism for improved feature extraction in the Feature Pyramid Network (FPN) and Path Aggregation Network (PAN). Moreover, it also employs the Alpha-IoU cost function for enhanced small-scale object detection. YOLOX-Ray's performance was assessed in three case studies: hotspot detection, infrastructure crack detection and corrosion detection. The architecture outperforms all other configurations, achieving m A P 50 values of 89%, 99.6% and 87.7%, respectively. For the most challenging metric, m A P 50 : 95 , the achieved values were 44.7%, 66.1% and 51.8%, respectively. A comparative analysis demonstrated the importance of combining the SimAM attention mechanism with Alpha-IoU loss function for optimal performance. In conclusion, YOLOX-Ray's ability to detect and to locate multi-scale objects in industrial environments presents new opportunities for effective, efficient and sustainable inspection processes across various industries, revolutionizing the field of industrial inspections. [ABSTRACT FROM AUTHOR]

Object detection is a fundamental task in computer vision. Over the past several years, convolutional neural network (CNN)-based object detection models have significantly improved detection accuracyin terms of average precision (AP). Furthermore, feature pyramid networks (FPNs) are essential modules for object detection models to consider various object scales. However, the AP for small objects is lower than the AP for medium and large objects. It is difficult to recognize small objects because they do not have sufficient information, and information is lost in deeper CNN layers. This paper proposes a new FPN model named ssFPN (scale sequence (S 2 ) feature-based feature pyramid network) to detect multi-scale objects, especially small objects. We propose a new scale sequence (S 2 ) feature that is extracted by 3D convolution on the level of the FPN. It is defined and extracted from the FPN to strengthen the information on small objects based on scale-space theory. Motivated by this theory, the FPN is regarded as a scale space and extracts a scale sequence (S 2 ) feature by three-dimensional convolution on the level axis of the FPN. The defined feature is basically scale-invariant and is built on a high-resolution pyramid feature map for small objects. Additionally, the deigned S 2 feature can be extended to most object detection models based on FPNs. We also designed a feature-level super-resolution approach to show the efficiency of the scale sequence (S 2 ) feature. We verified that the scale sequence (S 2 ) feature could improve the classification accuracy for low-resolution images by training a feature-level super-resolution model. To demonstrate the effect of the scale sequence (S 2 ) feature, experiments on the scale sequence (S 2 ) feature built-in object detection approach including both one-stage and two-stage models were conducted on the MS COCO dataset. For the two-stage object detection models Faster R-CNN and Mask R-CNN with the S 2 feature, AP improvements of up to 1.6% and 1.4%, respectively, were achieved. Additionally, the AP S of each model was improved by 1.2% and 1.1%, respectively. Furthermore, the one-stage object detection models in the YOLO series were improved. For YOLOv4-P5, YOLOv4-P6, YOLOR-P6, YOLOR-W6, and YOLOR-D6 with the S 2 feature, 0.9%, 0.5%, 0.5%, 0.1%, and 0.1% AP improvements were observed. For small object detection, the AP S increased by 1.1%, 1.1%, 0.9%, 0.4%, and 0.1%, respectively. Experiments using the feature-level super-resolution approach with the proposed scale sequence (S 2 ) feature were conducted on the CIFAR-100 dataset. By training the feature-level super-resolution model, we verified that ResNet-101 with the S 2 feature trained on LR images achieved a 55.2% classification accuracy, which was 1.6% higher than for ResNet-101 trained on HR images. [ABSTRACT FROM AUTHOR]

Blind people often encounter challenges in managing their clothing, specifically in identifying defects such as stains or holes. With the progress of the computer vision field, it is crucial to minimize these limitations as much as possible to assist blind people with selecting appropriate clothing. Therefore, the objective of this paper is to use object detection technology to categorize and detect stains on garments. The defect detection system proposed in this study relies on the You Only Look Once (YOLO) architecture, which is a single-stage object detector that is well-suited for automated inspection tasks. The authors collected a dataset of clothing with defects and used it to train and evaluate the proposed system. The methodology used for the optimization of the defect detection system was based on three main components: (i) increasing the dataset with new defects, illumination conditions, and backgrounds, (ii) introducing data augmentation, and (iii) introducing defect classification. The authors compared and evaluated three different YOLOv5 models. The results of this study demonstrate that the proposed approach is effective and suitable for different challenging defect detection conditions, showing high average precision (AP) values, and paving the way for a mobile application to be accessible for the blind community. [ABSTRACT FROM AUTHOR]

Gaze estimation is an established research problem in computer vision. It has various applications in real life, from human–computer interactions to health care and virtual reality, making it more viable for the research community. Due to the significant success of deep learning techniques in other computer vision tasks—for example, image classification, object detection, object segmentation, and object tracking—deep learning-based gaze estimation has also received more attention in recent years. This paper uses a convolutional neural network (CNN) for person-specific gaze estimation. The person-specific gaze estimation utilizes a single model trained for one individual user, contrary to the commonly-used generalized models trained on multiple people's data. We utilized only low-quality images directly collected from a standard desktop webcam, so our method can be applied to any computer system equipped with such a camera without additional hardware requirements. First, we used the web camera to collect a dataset of face and eye images. Then, we tested different combinations of CNN parameters, including the learning and dropout rates. Our findings show that building a person-specific eye-tracking model produces better results with a selection of good hyperparameters when compared to universal models that are trained on multiple users' data. In particular, we achieved the best results for the left eye with 38.20 MAE (Mean Absolute Error) in pixels, the right eye with 36.01 MAE, both eyes combined with 51.18 MAE, and the whole face with 30.09 MAE, which is equivalent to approximately 1.45 degrees for the left eye, 1.37 degrees for the right eye, 1.98 degrees for both eyes combined, and 1.14 degrees for full-face images. [ABSTRACT FROM AUTHOR]

Anomalous road manhole covers pose a potential risk to road safety in cities. In the development of smart cities, computer vision techniques use deep learning to automatically detect anomalous manhole covers to avoid these risks. One important problem is that a large amount of data are required to train a road anomaly manhole cover detection model. The number of anomalous manhole covers is usually small, which makes it a challenge to create training datasets quickly. To expand the dataset and improve the generalization of the model, researchers usually copy and paste samples from the original data to other data in order to achieve data augmentation. In this paper, we propose a new data augmentation method, which uses data that do not exist in the original dataset as samples to automatically select the pasting position of manhole cover samples and predict the transformation parameters via visual prior experience and perspective transformations, making it more accurately capture the actual shape of manhole covers on a road. Without using other data enhancement processes, our method raises the mean average precision (mAP) by at least 6.8 compared with the baseline model. [ABSTRACT FROM AUTHOR]

With the rapid advancement of deep learning theory and hardware device computing capacity, computer vision tasks, such as object detection and instance segmentation, have entered a revolutionary phase in recent years. As a result, extremely challenging integrated tasks, such as person search, might develop quickly. The majority of efficient network frameworks, such as Seq-Net, are based on Faster R-CNN. However, because of the parallel structure of Faster R-CNN, the performance of re-ID can be significantly impacted by the single-layer, low resolution, and occasionally overlooked check feature diagrams retrieved during pedestrian detection. To address these issues, this paper proposed a person search methodology based on an inception convolution and feature fusion module (IC-FFM) using Seq-Net (Sequential End-to-end Network) as the benchmark. First, we replaced the general convolution in ResNet-50 with the new inception convolution module (ICM), allowing the convolution operation to effectively and dynamically distribute various channels. Then, to improve the accuracy of information extraction, the feature fusion module (FFM) was created to combine multi-level information using various levels of convolution. Finally, Bounding Box regression was created using convolution and the double-head module (DHM), which considerably enhanced the accuracy of pedestrian retrieval by combining global and fine-grained information. Experiments on CHUK-SYSU and PRW datasets showed that our method has higher accuracy than Seq-Net. In addition, our method is simpler and can be easily integrated into existing two-stage frameworks. [ABSTRACT FROM AUTHOR]

Because of the development of image processing using cameras and the subsequent development of artificial intelligence technology, various fields have begun to develop. However, it is difficult to implement an image processing algorithm that requires a lot of calculations on a light board. This paper proposes a method using real-time deep learning object recognition algorithms in lightweight embedded boards. We have developed an algorithm suitable for lightweight embedded boards by appropriately using two deep neural network architectures. The first architecture requires small computational volumes, although it provides low accuracy. The second architecture uses large computational volumes and provides high accuracy. The area is determined using the first architecture, which processes semantic segmentation with relatively little computation. After masking the area using the more accurate deep learning architecture, object detection is implemented with improved accuracy, as the image is filtered by segmentation and the cases that have not been recognized by various variables, such as differentiation from the background, are excluded. OpenCV (Open source Computer Vision) is used to process input images in Python, and images are processed using an efficient neural network (ENet) and You Only Look Once (YOLO). By running this algorithm, the average error can be reduced by approximately 2.4 times, allowing for more accurate object detection. In addition, object recognition can be performed in real time for lightweight embedded boards, as a rate of about 4 FPS (frames per second) is achieved. [ABSTRACT FROM AUTHOR]

Object detection is a computer vision task that involves localisation and classification of objects in an image. Video data implicitly introduces several challenges, such as blur, occlusion and defocus, making video object detection more challenging in comparison to still image object detection, which is performed on individual and independent images. This paper tackles these challenges by proposing an attention-heavy framework for video object detection that aggregates the disentangled features extracted from individual frames. The proposed framework is a two-stage object detector based on the Faster R-CNN architecture. The disentanglement head integrates scale, spatial and task-aware attention and applies it to the features extracted by the backbone network across all the frames. Subsequently, the aggregation head incorporates temporal attention and improves detection in the target frame by aggregating the features of the support frames. These include the features extracted from the disentanglement network along with the temporal features. We evaluate the proposed framework using the ImageNet VID dataset and achieve a mean Average Precision (mAP) of 49.8 and 52.5 using the backbones of ResNet-50 and ResNet-101, respectively. The improvement in performance over the individual baseline methods validates the efficacy of the proposed approach. [ABSTRACT FROM AUTHOR]

The diffractive deep neural network (D2NN) can efficiently accomplish 2D object recognition based on rapid optical manipulation. Moreover, the multiple-view D2NN array (MDA) possesses the obvious advantage of being able to effectively achieve 3D object classification. At present, 3D target recognition should be performed in a high-speed and dynamic way. It should be invariant to the typical shifting, scaling, and rotating variance of targets in relatively complicated circumstances, which remains a shortcoming of optical neural network architectures. In order to efficiently recognize 3D targets based on the developed D2NN, a more robust MDA (mr-MDA) is proposed in this paper. Through utilizing a new training strategy to tackle several random disturbances introduced into the optical neural network system, a trained mr-MDA model constructed by us was numerically verified, demonstrating that the training strategy is able to dynamically recognize 3D objects in a relatively stable way. [ABSTRACT FROM AUTHOR]

In recent years, due to the advancements in machine learning, object detection has become a mainstream task in the computer vision domain. The first phase of object detection is to find the regions where objects can exist. With the improvements in deep learning, traditional approaches, such as sliding windows and manual feature selection techniques, have been replaced with deep learning techniques. However, object detection algorithms face a problem when performed in low light, challenging weather, and crowded scenes, similar to any other task. Such an environment is termed a challenging environment. This paper exploits pixel-level information to improve detection under challenging situations. To this end, we exploit the recently proposed hybrid task cascade network. This network works collaboratively with detection and segmentation heads at different cascade levels. We evaluate the proposed methods on three complex datasets of ExDark, CURE-TSD, and RESIDE, and achieve a mAP of 0.71, 0.52, and 0.43, respectively. Our experimental results assert the efficacy of the proposed approach. [ABSTRACT FROM AUTHOR]

The safe in-field operation of autonomous agricultural vehicles requires detecting all objects that pose a risk of collision. Current vision-based algorithms for object detection and classification are unable to detect unknown classes of objects. In this paper, the problem is posed as anomaly detection instead, where convolutional autoencoders are applied to identify any objects deviating from the normal pattern. Training an autoencoder network to reconstruct normal patterns in agricultural fields makes it possible to detect unknown objects by high reconstruction error. Basic autoencoder (AE), vector-quantized variational autoencoder (VQ-VAE), denoising autoencoder (DAE) and semisupervised autoencoder (SSAE) with a max-margin-inspired loss function are investigated and compared with a baseline object detector based on YOLOv5. Results indicate that SSAE with an area under the curve for precision/recall (PR AUC) of 0.9353 outperforms other autoencoder models and is comparable to an object detector with a PR AUC of 0.9794. Qualitative results show that SSAE is capable of detecting unknown objects, whereas the object detector is unable to do so and fails to identify known classes of objects in specific cases. [ABSTRACT FROM AUTHOR]

Roads make a huge contribution to the economy and act as a platform for transportation. Potholes in roads are one of the major concerns in transportation infrastructure. A lot of research has proposed using computer vision techniques to automate pothole detection that include a wide range of image processing and object detection algorithms. There is a need to automate the pothole detection process with adequate accuracy and speed and implement the process easily and with low setup cost. In this paper, we have developed efficient deep learning convolution neural networks (CNNs) to detect potholes in real-time with adequate accuracy. To reduce the computational cost and improve the training results, this paper proposes a modified VGG16 (MVGG16) network by removing some convolution layers and using different dilation rates. Moreover, this paper uses the MVGG16 as a backbone network for the Faster R-CNN. In addition, this work compares the performance of YOLOv5 (Large (Yl), Medium (Ym), and Small (Ys)) models with ResNet101 backbone and Faster R-CNN with ResNet50(FPN), VGG16, MobileNetV2, InceptionV3, and MVGG16 backbones. The experimental results show that the Ys model is more applicable for real-time pothole detection because of its speed. In addition, using the MVGG16 network as the backbone of the Faster R-CNN provides better mean precision and shorter inference time than using VGG16, InceptionV3, or MobilNetV2 backbones. The proposed MVGG16 succeeds in balancing the pothole detection accuracy and speed. [ABSTRACT FROM AUTHOR]

In terms of small objects in traffic scenes, general object detection algorithms have low detection accuracy, high model complexity, and slow detection speed. To solve the above problems, an improved algorithm (named YOLO-MXANet) is proposed in this paper. Complete-Intersection over Union (CIoU) is utilized to improve loss function for promoting the positioning accuracy of the small object. In order to reduce the complexity of the model, we present a lightweight yet powerful backbone network (named SA-MobileNeXt) that incorporates channel and spatial attention. Our approach can extract expressive features more effectively by applying the Shuffle Channel and Spatial Attention (SCSA) module into the SandGlass Block (SGBlock) module while increasing the parameters by a small number. In addition, the data enhancement method combining Mosaic and Mixup is employed to improve the robustness of the training model. The Multi-scale Feature Enhancement Fusion (MFEF) network is proposed to fuse the extracted features better. In addition, the SiLU activation function is utilized to optimize the Convolution-Batchnorm-Leaky ReLU (CBL) module and the SGBlock module to accelerate the convergence of the model. The ablation experiments on the KITTI dataset show that each improved method is effective. The improved algorithm reduces the complexity and detection speed of the model while improving the object detection accuracy. The comparative experiments on the KITTY dataset and CCTSDB dataset with other algorithms show that our algorithm also has certain advantages. [ABSTRACT FROM AUTHOR]

With the introduction of deep learning, a significant amount of research has been conducted in the field of computer vision in the past decade. In particular, research on object detection (OD) continues to progress rapidly. However, despite these advances, some limitations need to be overcome to enable real-world applications of deep learning-based OD models. One such limitation is inaccurate OD when image quality is poor or a target object is small. The performance degradation phenomenon for small objects is similar to the fundamental limitations of an OD model, such as the constraint of the receptive field, which is a difficult problem to solve using only an OD model. Therefore, OD performance can be hindered by low image quality or small target objects. To address this issue, this study investigates the compatibility of super-resolution (SR) and OD techniques to improve detection, particularly for small objects. We analyze the combination of SR and OD models, classifying them based on architectural characteristics. The experimental results show a substantial improvement when integrating OD detectors with SR models. Overall, it was demonstrated that, when the evaluation metrics (PSNR, SSIM) of the SR models are high, the performance in OD is correspondingly high as well. Especially, evaluations on the MS COCO dataset reveal that the enhancement rate for small objects is 9.4% higher compared to all objects. This work provides an analysis of SR and OD model compatibility, demonstrating the potential benefits of their synergistic combination. The experimental code can be found on our GitHub repository. [ABSTRACT FROM AUTHOR]

LiDAR-based 3D object detection and localization are crucial components of autonomous navigation systems, including autonomous vehicles and mobile robots. Most existing LiDAR-based 3D object detection and localization approaches primarily use geometric or structural feature abstractions from LiDAR point clouds. However, these approaches can be susceptible to environmental noise due to adverse weather conditions or the presence of highly scattering media. In this work, we propose an intensity-aware voxel encoder for robust 3D object detection. The proposed voxel encoder generates an intensity histogram that describes the distribution of point intensities within a voxel and is used to enhance the voxel feature set. We integrate this intensity-aware encoder into an efficient single-stage voxel-based detector for 3D object detection. Experimental results obtained using the KITTI dataset show that our method achieves comparable results with respect to the state-of-the-art method for car objects in 3D detection and from a bird's-eye view and superior results for pedestrian and cyclic objects. Furthermore, our model can achieve a detection rate of 40.7 FPS during inference time, which is higher than that of the state-of-the-art methods and incurs a lower computational cost. [ABSTRACT FROM AUTHOR]

Autonomous Vehicles (AVs) have the potential to solve many traffic problems, such as accidents, congestion and pollution. However, there are still challenges to overcome, for instance, AVs need to accurately perceive their environment to safely navigate in busy urban scenarios. The aim of this paper is to review recent articles on computer vision techniques that can be used to build an AV perception system. AV perception systems need to accurately detect non-static objects and predict their behaviour, as well as to detect static objects and recognise the information they are providing. This paper, in particular, focuses on the computer vision techniques used to detect pedestrians and vehicles. There have been many papers and reviews on pedestrians and vehicles detection so far. However, most of the past papers only reviewed pedestrian or vehicle detection separately. This review aims to present an overview of the AV systems in general, and then review and investigate several detection computer vision techniques for pedestrians and vehicles. The review concludes that both traditional and Deep Learning (DL) techniques have been used for pedestrian and vehicle detection; however, DL techniques have shown the best results. Although good detection results have been achieved for pedestrians and vehicles, the current algorithms still struggle to detect small, occluded, and truncated objects. In addition, there is limited research on how to improve detection performance in difficult light and weather conditions. Most of the algorithms have been tested on well-recognised datasets such as Caltech and KITTI; however, these datasets have their own limitations. Therefore, this paper recommends that future works should be implemented on more new challenging datasets, such as PIE and BDD100K. [ABSTRACT FROM AUTHOR]

Several studies in computer vision have examined specular removal, which is crucial for object detection and recognition. This research has traditionally been divided into two tasks: specular highlight removal, which focuses on removing specular highlights on object surfaces, and reflection removal, which deals with specular reflections occurring on glass surfaces. In reality, however, both types of specular effects often coexist, making it a fundamental challenge that has not been adequately addressed. Recognizing the necessity of integrating specular components handled in both tasks, we constructed a specular-light (S-Light) DB for training single-image-based deep learning models. Moreover, considering the absence of benchmark datasets for quantitative evaluation, the multi-scale normalized cross correlation (MS-NCC) metric, which considers the correlation between specular and diffuse components, was introduced to assess the learning outcomes. [ABSTRACT FROM AUTHOR]

Deep learning has rapidly increased in popularity, leading to the development of perception solutions for autonomous driving. The latter field leverages techniques developed for computer vision in other domains for accomplishing perception tasks such as object detection. However, the black-box nature of deep neural models and the complexity of the autonomous driving context motivates the study of explainability in these models that perform perception tasks. Moreover, this work explores explainable AI techniques for the object detection task in the context of autonomous driving. An extensive and detailed comparison is carried out between gradient-based and perturbation-based methods (e.g., D-RISE). Moreover, several experimental setups are used with different backbone architectures and different datasets to observe the influence of these aspects in the explanations. All the techniques explored consist of saliency methods, making their interpretation and evaluation primarily visual. Nevertheless, numerical assessment methods are also used. Overall, D-RISE and guided backpropagation obtain more localized explanations. However, D-RISE highlights more meaningful regions, providing more human-understandable explanations. To the best of our knowledge, this is the first approach to obtaining explanations focusing on the regression of the bounding box coordinates. [ABSTRACT FROM AUTHOR]

The recent large-scale fire incidents on construction sites in South Korea have highlighted the need for computer vision technology to detect fire risks before an actual occurrence of fire. This study developed a proactive fire risk detection system by detecting the coexistence of an ignition source (sparks) and a combustible material (urethane foam or Styrofoam) using object detection on images from a surveillance camera. Statistical analysis was carried out on fire incidences on construction sites in South Korea to provide insight into the cause of the large-scale fire incidents. Labeling approaches were discussed to improve the performance of the object detectors for sparks and urethane foams. Detecting ignition sources and combustible materials at a distance was discussed in order to improve the performance for long-distance objects. Two candidate deep learning models, Yolov5 and EfficientDet, were compared in their performance. It was found that Yolov5 showed slightly higher mAP performances: Yolov5 models showed mAPs from 87% to 90% and EfficientDet models showed mAPs from 82% to 87%, depending on the complexity of the model. However, Yolov5 showed distinctive advantages over EfficientDet in terms of easiness and speed of learning. [ABSTRACT FROM AUTHOR]

In multi-cat households, monitoring individual cats' various behaviors is essential for diagnosing their health and ensuring their well-being. This study focuses on the defecation and urination activities of cats, and introduces an adaptive cat identification architecture based on deep learning (DL) and machine learning (ML) methods. The architecture comprises an object detector and a classification module, with the primary focus on the design of the classification component. The DL object detection algorithm, YOLOv4, is used for the cat object detector, with the convolutional neural network, EfficientNetV2, serving as the backbone for our feature extractor in identity classification with several ML classifiers. Additionally, to address changes in cat composition and individual cat appearances in multi-cat households, we propose an adaptive concept drift approach involving retraining the classification module. To support our research, we compile a comprehensive cat body dataset comprising 8934 images of 36 cats. After a rigorous evaluation of different combinations of DL models and classifiers, we find that the support vector machine (SVM) classifier yields the best performance, achieving an impressive identification accuracy of 94.53%. This outstanding result underscores the effectiveness of the system in accurately identifying cats. [ABSTRACT FROM AUTHOR]

Too often, the testing and evaluation of object detection, as well as the classification techniques for high-resolution remote sensing imagery, are confined to clean, discretely partitioned datasets, i.e., the closed-world model. In recent years, the performance on a number of benchmark datasets has exceeded 99% when evaluated using cross-validation techniques. However, real-world remote sensing data are truly big data, which often exceed billions of pixels. Therefore, one of the greatest challenges regarding the evaluation of machine learning models taken out of the clean laboratory setting and into the real world is the difficulty of measuring performance. It is necessary to evaluate these models on a grander scale, namely, tens of thousands of square kilometers, where it is intractable to the ground truth and the ever-changing anthropogenic surface of Earth. The ultimate goal of computer vision model development for automated analysis and broad area search and discovery is to augment and assist humans, specifically human–machine teaming for real-world tasks. In this research, various models have been trained using object classes from benchmark datasets such as UC Merced, PatternNet, RESISC-45, and MDSv2. We detail techniques to scan broad swaths of the Earth with deep convolutional neural networks. We present algorithms for localizing object detection results, as well as a methodology for the evaluation of the results of broad-area scans. Our research explores the challenges of transitioning these models out of the training–validation laboratory setting and into the real-world application domain. We show a scalable approach to leverage state-of-the-art deep convolutional neural networks for the search, detection, and annotation of objects within large swaths of imagery, with the ultimate goal of providing a methodology for evaluating object detection machine learning models in real-world scenarios. [ABSTRACT FROM AUTHOR]

For facing of the problems caused by the YOLOv4 algorithm's insensitivity to small objects and low detection precision in traffic light detection and recognition, the Improved YOLOv4 algorithm is investigated in the paper using the shallow feature enhancement mechanism and the bounding box uncertainty prediction mechanism. The shallow feature enhancement mechanism is used to extract features from the network and improve the network's ability to locate small objects and color resolution by merging two shallow features at different stages with the high-level semantic features obtained after two rounds of upsampling. Uncertainty is introduced in the bounding box prediction mechanism to improve the reliability of the prediction of the bounding box by modeling the output coordinates of the prediction bounding box and adding the Gaussian model to calculate the uncertainty of the coordinate information. The LISA traffic light data set is used to perform detection and recognition experiments separately. The Improved YOLOv4 algorithm is shown to have a high effectiveness in enhancing the detection and recognition precision of traffic lights. In the detection experiment, the area under the PR curve value of the Improved YOLOv4 algorithm is found to be 97.58%, which represents an increase of 7.09% in comparison to the 90.49% score gained in the Vision for Intelligent Vehicles and Applications Challenge Competition. In the recognition experiment, the mean average precision of the Improved YOLOv4 algorithm is 82.15%, which is 2.86% higher than that of the original YOLOv4 algorithm. The Improved YOLOv4 algorithm shows remarkable advantages as a robust and practical method for use in the real-time detection and recognition of traffic signal lights. [ABSTRACT FROM AUTHOR]

This paper focuses on improving the performance of scientific instrumentation that uses glass spray chambers for sample introduction, such as spectrometers, which are widely used in analytical chemistry, by detecting incidents using deep convolutional models. The performance of these instruments can be affected by the quality of the introduction of the sample into the spray chamber. Among the indicators of poor quality sample introduction are two primary incidents: The formation of liquid beads on the surface of the spray chamber, and flooding at the bottom of the spray chamber. Detecting such events autonomously as they occur can assist with improving the overall operational accuracy and efficacy of the chemical analysis, and avoid severe incidents such as malfunction and instrument damage. In contrast to objects commonly seen in the real world, beading and flooding detection are more challenging since they are of significantly small size and transparent. Furthermore, the non-rigid property increases the difficulty of the detection of these incidents, as such that existing deep-learning-based object detection frameworks are prone to fail for this task. There is no former work that uses computer vision to detect these incidents in the chemistry industry. In this work, we propose two frameworks for the detection task of these two incidents, which not only leverage the modern deep learning architectures but also integrate with expert knowledge of the problems. Specifically, the proposed networks first localize the regions of interest where the incidents are most likely generated and then refine these incident outputs. The use of data augmentation and synthesis, and choice of negative sampling in training, allows for a large increase in accuracy while remaining a real-time system for inference. In the data collected from our laboratory, our method surpasses widely used object detection baselines and can correctly detect 95% of the beads and 98% of the flooding. At the same time, out method can process four frames per second and is able to be implemented in real time. [ABSTRACT FROM AUTHOR]

Submarines are considered extremely strategic for any naval army due to their stealth capability. Periscopes are crucial sensors for these vessels, and emerging to the surface or periscope depth is required to identify visual contacts through this device. This maneuver has many procedures and usually has to be fast and agile to avoid exposure. This paper presents and implements a novel architecture for real submarine periscopes developed for future Brazilian naval fleet operations. Our system consists of a probe that is connected to the craft and carries a 360 camera. We project and take the images inside the vessel using traditional VR/XR devices. We also propose and implement an efficient computer vision-based MR technique to estimate and display detected vessels effectively and precisely. The vessel detection model is trained using synthetic images. So, we built and made available a dataset composed of 99,000 images. Finally, we also estimate distances of the classified elements, showing all the information in an AR-based interface. Although the probe is wired-connected, it allows for the vessel to stand in deep positions, reducing its exposure and introducing a new way for submarine maneuvers and operations. We validate our proposal through a user experience experiment using 19 experts in periscope operations. [ABSTRACT FROM AUTHOR]

Video analytics and computer vision applications face challenges when using video sequences with low visibility. The visibility of a video sequence is degraded when the sequence is affected by atmospheric interference like rain. Many approaches have been proposed to remove rain streaks from video sequences. Some approaches are based on physical features, and some are based on data-driven (i.e., deep-learning) models. Although the physical features-based approaches have better rain interpretability, the challenges are extracting the appropriate features and fusing them for meaningful rain removal, as the rain streaks and moving objects have dynamic physical characteristics and are difficult to distinguish. Additionally, the outcome of the data-driven models mostly depends on variations relating to the training dataset. It is difficult to include datasets with all possible variations in model training. This paper addresses both issues and proposes a novel hybrid technique where we extract novel physical features and data-driven features and then combine them to create an effective rain-streak removal strategy. The performance of the proposed algorithm has been tested in comparison to several relevant and contemporary methods using benchmark datasets. The experimental result shows that the proposed method outperforms the other methods in terms of subjective, objective, and object detection comparisons for both synthetic and real rain scenarios by removing rain streaks and retaining the moving objects more effectively. [ABSTRACT FROM AUTHOR]

Improving soybean (Glycine max L. (Merr.)) yield is crucial for strengthening national food security. Predicting soybean yield is essential to maximize the potential of crop varieties. Non-destructive methods are needed to estimate yield before crop maturity. Various approaches, including the pod-count method, have been used to predict soybean yield, but they often face issues with the crop background color. To address this challenge, we explored the application of a depth camera to real-time filtering of RGB images, aiming to enhance the performance of the pod-counting classification model. Additionally, this study aimed to compare object detection models (YOLOV7 and YOLOv7-E6E) and select the most suitable deep learning (DL) model for counting soybean pods. After identifying the best architecture, we conducted a comparative analysis of the model's performance by training the DL model with and without background removal from images. Results demonstrated that removing the background using a depth camera improved YOLOv7's pod detection performance by 10.2% precision, 16.4% recall, 13.8% mAP@50, and 17.7% mAP@0.5:0.95 score compared to when the background was present. Using a depth camera and the YOLOv7 algorithm for pod detection and counting yielded a mAP@0.5 of 93.4% and mAP@0.5:0.95 of 83.9%. These results indicated a significant improvement in the DL model's performance when the background was segmented, and a reasonably larger dataset was used to train YOLOv7. [ABSTRACT FROM AUTHOR]

The development of an accurate and efficient method for detecting missing bolts in engineering structures is crucial. To this end, a missing bolt detection method that leveraged machine vision and deep learning was developed. First, a comprehensive dataset of bolt images captured under natural conditions was constructed, which improved the generality and recognition accuracy of the trained bolt target detection model. Second, three deep learning network models, namely, YOLOv4, YOLOv5s, and YOLOXs, were compared, and YOLOv5s was selected as the bolt target detection model. With YOLOv5s as the target recognition model, the bolt head and bolt nut had average precisions of 0.93 and 0.903, respectively. Third, a missing bolt detection method based on perspective transformation and IoU was presented and validated under laboratory conditions. Finally, the proposed method was applied to an actual footbridge structure to test its feasibility and effectiveness in real engineering scenarios. The experimental results showed that the proposed method could accurately identify bolt targets with a confidence level of over 80% and detect missing bolts under different image distances, perspective angles, light intensities, and image resolutions. Moreover, the experimental results on a footbridge demonstrated that the proposed method could reliably detect the missing bolt even at a shooting distance of 1 m. The proposed method provided a low-cost, efficient, and automated technical solution for the safety management of bolted connection components in engineering structures. [ABSTRACT FROM AUTHOR]

Hundreds of people are injured or killed in road accidents. These accidents are caused by several intrinsic and extrinsic factors, including the attentiveness of the driver towards the road and its associated features. These features include approaching vehicles, pedestrians, and static fixtures, such as road lanes and traffic signs. If a driver is made aware of these features in a timely manner, a huge chunk of these accidents can be avoided. This study proposes a computer vision-based solution for detecting and recognizing traffic types and signs to help drivers pave the door for self-driving cars. A real-world roadside dataset was collected under varying lighting and road conditions, and individual frames were annotated. Two deep learning models, YOLOv7 and Faster RCNN, were trained on this custom-collected dataset to detect the aforementioned road features. The models produced mean Average Precision (mAP) scores of 87.20% and 75.64%, respectively, along with class accuracies of over 98.80%; all of these were state-of-the-art. The proposed model provides an excellent benchmark to build on to help improve traffic situations and enable future technological advances, such as Advance Driver Assistance System (ADAS) and self-driving cars. [ABSTRACT FROM AUTHOR]

This research describes the use of high-performance computing (HPC) and deep learning to create prediction models that could be deployed on edge AI devices equipped with camera and installed in poultry farms. The main idea is to leverage an existing IoT farming platform and use HPC offline to run deep learning to train the models for object detection and object segmentation, where the objects are chickens in images taken on farm. The models can be ported from HPC to edge AI devices to create a new type of computer vision kit to enhance the existing digital poultry farm platform. Such new sensors enable implementing functions such as counting chickens, detection of dead chickens, and even assessing their weight or detecting uneven growth. These functions combined with the monitoring of environmental parameters, could enable early disease detection and improve the decision-making process. The experiment focused on Faster R-CNN architectures and AutoML was used to identify the most suitable architecture for chicken detection and segmentation for the given dataset. For the selected architectures, further hyperparameter optimization was carried out and we achieved the accuracy of AP = 85%, AP50 = 98%, and AP75 = 96% for object detection and AP = 90%, AP50 = 98%, and AP75 = 96% for instance segmentation. These models were installed on edge AI devices and evaluated in the online mode on actual poultry farms. Initial results are promising, but further development of the dataset and improvements in prediction models is needed. [ABSTRACT FROM AUTHOR]

Activity recognition in unmanned aerial vehicle (UAV) surveillance is addressed in various computer vision applications such as image retrieval, pose estimation, object detection, object detection in videos, object detection in still images, object detection in video frames, face recognition, and video action recognition. In the UAV-based surveillance technology, video segments captured from aerial vehicles make it challenging to recognize and distinguish human behavior. In this research, to recognize a single and multi-human activity using aerial data, a hybrid model of histogram of oriented gradient (HOG), mask-regional convolutional neural network (Mask-RCNN), and bidirectional long short-term memory (Bi-LSTM) is employed. The HOG algorithm extracts patterns, Mask-RCNN extracts feature maps from the raw aerial image data, and the Bi-LSTM network exploits the temporal relationship between the frames for the underlying action in the scene. This Bi-LSTM network reduces the error rate to the greatest extent due to its bidirectional process. This novel architecture generates enhanced segmentation by utilizing the histogram gradient-based instance segmentation and improves the accuracy of classifying human activities using the Bi-LSTM approach. Experimental outcomes demonstrate that the proposed model outperforms the other state-of-the-art models and has achieved 99.25% accuracy on the YouTube-Aerial dataset. [ABSTRACT FROM AUTHOR]

Deep learning-based computer vision algorithms, especially image segmentation, have been successfully applied to pixel-level crack detection. The prediction accuracy relies heavily on detecting the performance of fine-grained cracks and removing crack-like noise. We propose a fast encoder-decoder network with scaling attention. We focus on a low-level feature map by minimizing encoder-decoder pairs and adopting an Atrous Spatial Pyramid Pooling (ASPP) layer to improve the detection accuracy of tiny cracks. Another challenge is the reduction in crack-like noise. This introduces a novel scaling attention, A G + , to suppress irrelevant regions. However, removing crack-like noise, such as grooving, is difficult by using only improved segmentation networks. In this study, a crack dataset is generated. It contains 11,226 sets of images and masks, which are effective for detecting detailed tiny cracks and removing non-semantic objects. Our model is evaluated on the generated dataset and compared with state-of-the-art segmentation networks. We use the mean Dice coefficient (mDice) and mean Intersection over union (mIoU) to compare the performance and FLOPs for computational complexity. The experimental results show that our model improves the detection accuracy of fine-grained cracks and reduces the computational cost dramatically. The mDice score of the proposed model is close to the best score, with only a 1.2% difference but two times fewer FLOPs. [ABSTRACT FROM AUTHOR]

In order to save manpower on rail track inspection, computer vision-based methodologies are developed. We propose utilizing the YOLOv4-Tiny neural network to identify track defects in real time. There are ten defects covering fasteners, rail surfaces, and sleepers from the upward and six defects about the rail waist from the sideward. The proposed real-time inspection system includes a high-performance notebook, two sports cameras, and three parallel processes. The hardware is mounted on a flat cart running at 30 km/h. The inspection results about the abnormal track components could be queried by defective type, time, and the rail hectometer stake. In the experiments, data augmentation by a Cycle Generative Adversarial Network (GAN) is used to increase the dataset. The number of images is 3800 on the upward and 967 on the sideward. Five object detection neural network models—YOLOv4, YOLOv4-Tiny, YOLOX-Tiny, SSD512, and SSD300—were tested. The YOLOv4-Tiny model with 150 FPS is selected as the recognition kernel, as it achieved 91.7%, 92%, and 91% for the mAP, precision, and recall of the defective track components from the upward, respectively. The mAP, precision, and recall of the defective track components from the sideward are 99.16%, 96%, and 94%, respectively. [ABSTRACT FROM AUTHOR]

The utilization of computer vision in smart farming is becoming a trend in constructing an agricultural automation scheme. Deep learning (DL) is famous for the accurate approach to addressing the tasks in computer vision, such as object detection and image classification. The superiority of the deep learning model on the smart farming application, called Progressive Contextual Excitation Network (PCENet), has also been studied in our recent study to classify cocoa bean images. However, the assessment of the computational time on the PCENet model shows that the original model is only 0.101s or 9.9 FPS on the Jetson Nano as the edge platform. Therefore, this research demonstrates the compression technique to accelerate the PCENet model using pruning filters. From our experiment, we can accelerate the current model and achieve 16.7 FPS assessed in the Jetson Nano. Moreover, the accuracy of the compressed model can be maintained at 86.1%, while the original model is 86.8%. In addition, our approach is more accurate than ResNet18 as the state-of-the-art only reaches 82.7%. The assessment using the corn leaf disease dataset indicates that the compressed model can achieve an accuracy of 97.5%, while the accuracy of the original PCENet is 97.7%. [ABSTRACT FROM AUTHOR]

LiDAR is a commonly used sensor for autonomous driving to make accurate, robust, and fast decision-making when driving. The sensor is used in the perception system, especially object detection, to understand the driving environment. Although 2D object detection has succeeded during the deep-learning era, the lack of depth information limits understanding of the driving environment and object location. Three-dimensional sensors, such as LiDAR, give 3D information about the surrounding environment, which is essential for a 3D perception system. Despite the attention of the computer vision community to 3D object detection due to multiple applications in robotics and autonomous driving, there are challenges, such as scale change, sparsity, uneven distribution of LiDAR data, and occlusions. Different representations of LiDAR data and methods to minimize the effect of the sparsity of LiDAR data have been proposed. This survey presents the LiDAR-based 3D object detection and feature-extraction techniques for LiDAR data. The 3D coordinate systems differ in camera and LiDAR-based datasets and methods. Therefore, the commonly used 3D coordinate systems are summarized. Then, state-of-the-art LiDAR-based 3D object-detection methods are reviewed with a selected comparison among methods. [ABSTRACT FROM AUTHOR]

Features play a critical role in computer vision tasks. Deep learning methods have resulted in significant breakthroughs in the field of object detection, but it is still an extremely challenging obstacle when an object is very small. In this work, we propose a feature-enhancement- and channel-attention-guided single-shot detector called the FCSSD with four modules to improve object detection performance. Specifically, inspired by the structure of atrous convolution, we built an efficient feature-extraction module (EFM) in order to explore contextual information along the spatial dimension, and then pyramidal aggregation module (PAM) is presented to explore the semantic features of deep layers, thus reducing the semantic gap between multi-scale features. Furthermore, we construct an effective feature pyramid refinement fusion (FPRF) to refine the multi-scale features and create benefits for richer object knowledge. Finally, an attention-guided module (AGM) is developed to balance the channel weights and optimize the final integrated features on each level; this alleviates the aliasing effects of the FPN with negligible computational costs. The FCSSD exploits richer information of shallow layers and higher layers by using our designed modules, thus accomplishing excellent detection performance for multi-scale object detection and reaching a better tradeoff between accuracy and inference time. Experiments on PASCAL VOC and MS COCO datasets were conducted to evaluate the performance, showing that our FCSSD achieves competitive detection performance compared with existing mainstream object detection methods. [ABSTRACT FROM AUTHOR]

Object detection is an essential capability for performing complex tasks in robotic applications. Today, deep learning (DL) approaches are the basis of state-of-the-art solutions in computer vision, where they provide very high accuracy albeit with high computational costs. Due to the physical limitations of robotic platforms, embedded devices are not as powerful as desktop computers, and adjustments have to be made to deep learning models before transferring them to robotic applications. This work benchmarks deep learning object detection models in embedded devices. Furthermore, some hardware selection guidelines are included, together with a description of the most relevant features of the two boards selected for this benchmark. Embedded electronic devices integrate a powerful AI co-processor to accelerate DL applications. To take advantage of these co-processors, models must be converted to a specific embedded runtime format. Five quantization levels applied to a collection of DL models are considered; two of them allow the execution of models in the embedded general-purpose CPU and are used as the baseline to assess the improvements obtained when running the same models with the three remaining quantization levels in the AI co-processors. The benchmark procedure is explained in detail, and a comprehensive analysis of the collected data is presented. Finally, the feasibility and challenges of the implementation of embedded object detection applications are discussed. [ABSTRACT FROM AUTHOR]

With the development of neural networks, object detection based on deep learning is developing rapidly, and its applications are gradually increasing. In the tire industry, detecting speckle interference bubble defects of tire crown has difficulties such as low image contrast, small object scale, and large internal differences of defects, which affect the detection precision. To solve these problems, we propose a new feature pyramid network based on Faster RCNN-FPN. It can fuse features across levels and directions to improve small object detection and localization, and increase object detection precision. The method has proven its effectiveness through cross-validation experiments. On a tire crown bubble defect dataset, the mAP [0.5:0.95] increased by 2.08% and the AP0.5 increased by 2.4% over the original network. The results show that the improved network significantly improves detecting tire crown bubble defects. [ABSTRACT FROM AUTHOR]

Object detection is a vital step in satellite imagery-based computer vision applications such as precision agriculture, urban planning and defense applications. In satellite imagery, object detection is a very complicated task due to various reasons including low pixel resolution of objects and detection of small objects in the large scale (a single satellite image taken by Digital Globe comprises over 240 million pixels) satellite images. Object detection in satellite images has many challenges such as class variations, multiple objects pose, high variance in object size, illumination and a dense background. This study aims to compare the performance of existing deep learning algorithms for object detection in satellite imagery. We created the dataset of satellite imagery to perform object detection using convolutional neural network-based frameworks such as faster RCNN (faster region-based convolutional neural network), YOLO (you only look once), SSD (single-shot detector) and SIMRDWN (satellite imagery multiscale rapid detection with windowed networks). In addition to that, we also performed an analysis of these approaches in terms of accuracy and speed using the developed dataset of satellite imagery. The results showed that SIMRDWN has an accuracy of 97% on high-resolution images, while Faster RCNN has an accuracy of 95.31% on the standard resolution (1000 × 600). YOLOv3 has an accuracy of 94.20% on standard resolution (416 × 416) while on the other hand SSD has an accuracy of 84.61% on standard resolution (300 × 300). When it comes to speed and efficiency, YOLO is the obvious leader. In real-time surveillance, SIMRDWN fails. When YOLO takes 170 to 190 milliseconds to perform a task, SIMRDWN takes 5 to 103 milliseconds. [ABSTRACT FROM AUTHOR]

Limited training data is one of the biggest challenges in the industrial application of deep learning. Generating synthetic training images is a promising solution in computer vision; however, minimizing the domain gap between synthetic and real-world images remains a problem. Therefore, based on a real-world application, we explored the generation of images with physics-based rendering for an industrial object detection task. Setting up the render engine's environment requires a lot of choices and parameters. One fundamental question is whether to apply the concept of domain randomization or use domain knowledge to try and achieve photorealism. To answer this question, we compared different strategies for setting up lighting, background, object texture, additional foreground objects and bounding box computation in a data-centric approach. We compared the resulting average precision from generated images with different levels of realism and variability. In conclusion, we found that domain randomization is a viable strategy for the detection of industrial objects. However, domain knowledge can be used for object-related aspects to improve detection performance. Based on our results, we provide guidelines and an open-source tool for the generation of synthetic images for new industrial applications. [ABSTRACT FROM AUTHOR]

Object detection, classification and tracking are three important computer vision techniques. Object detection, classification and tracking represent a collection of computer vision tasks that are far from completely solved. Training data curation is an important issue in deep learning research because the performance of the deep learning model is heavily dependent on the size of the training dataset. [Extracted from the article]

We present an end-to-end smart harvesting solution for precision agriculture. Our proposed pipeline begins with yield estimation that is done through the use of object detection and tracking to count fruit within a video. We use and train You Only Look Once model (YOLO) on video clips of apples, oranges and pumpkins. The bounding boxes obtained through objection detection are used as an input to our selected tracking model, DeepSORT. The original version of DeepSORT is unusable with fruit data, as the appearance feature extractor only works with people. We implement ResNet as DeepSORT's new feature extractor, which is lightweight, accurate and generically works on different fruits. Our yield estimation module shows accuracy between 91–95% on real footage of apple trees. Our modification successfully works for counting oranges and pumpkins, with an accuracy of 79% and 93.9% with no need for training. Our framework additionally includes a visualization of the yield. This is done through the incorporation of geospatial data. We also propose a mechanism to annotate a set of frames with a respective GPS coordinate. During counting, the count within the set of frames and the matching GPS coordinate are recorded, which we then visualize on a map. We leverage this information to propose an optimal container placement solution. Our proposed solution involves minimizing the number of containers to place across the field before harvest, based on a set of constraints. This acts as a decision support system for the farmer to make efficient plans for logistics, such as labor, equipment and gathering paths before harvest. Our work serves as a blueprint for future agriculture decision support systems that can aid in many other aspects of farming. [ABSTRACT FROM AUTHOR]