Search

Object detection is one of the core technologies for autonomous driving. Current road object detection mainly relies on visible light, which is prone to missed detections and false alarms in rainy, night-time, and foggy scenes. Multispectral object detection based on the fusion of RGB and infrared images can effectively address the challenges of complex and changing road scenes, improving the detection performance of current algorithms in complex scenarios. However, previous multispectral detection algorithms suffer from issues such as poor fusion of dual-mode information, poor detection performance for multi-scale objects, and inadequate utilization of semantic information. To address these challenges and enhance the detection performance in complex road scenes, this paper proposes a novel multispectral object detection algorithm called MRD-YOLO. In MRD-YOLO, we utilize interaction-based feature extraction to effectively fuse information and introduce the BIC-Fusion module with attention guidance to fuse different modal information. We also incorporate the SAConv module to improve the model's detection performance for multi-scale objects and utilize the AIFI structure to enhance the utilization of semantic information. Finally, we conduct experiments on two major public datasets, FLIR_Aligned and M3FD. The experimental results demonstrate that compared to other algorithms, the proposed algorithm achieves superior detection performance in complex road scenes. [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]

The purpose of image dehazing is to remove the interference from weather factors in degraded images and enhance the clarity and color saturation of images to maximize the restoration of useful features. Single image dehazing is one of the most important tasks in the field of image restoration. In recent years, due to the progress of deep learning, single image dehazing has made great progress. With the success of Transformer in advanced computer vision tasks, some research studies also began to apply Transformer to image dehazing tasks and obtained surprising results. However, both the deconvolution-neural-network-based dehazing algorithm and Transformer based dehazing algorithm magnify their advantages and disadvantages separately. Therefore, this paper proposes a novel Transformer–Convolution fusion dehazing network (TCFDN), which uses Transformer's global modeling ability and convolutional neural network's local modeling ability to improve the dehazing ability. In the Transformer–Convolution fusion dehazing network, the classic self-encoder structure is used. This paper proposes a Transformer–Convolution hybrid layer, which uses an adaptive fusion strategy to make full use of the Swin-Transformer and convolutional neural network to extract and reconstruct image features. On the basis of previous research, this layer further improves the ability of the network to remove haze. A series of contrast experiments and ablation experiments not only proved that the Transformer–Convolution fusion dehazing network proposed in this paper exceeded the more advanced dehazing algorithm, but also provided solid and powerful evidence for the basic theory on which it depends. [ABSTRACT FROM AUTHOR]

Sugarcane stem node identification is the core technology required for the intelligence and mechanization of the sugarcane industry. However, detecting stem nodes quickly and accurately is still a significant challenge. In this paper, in order to solve this problem, a new algorithm combining YOLOv3 and traditional methods of computer vision is proposed, which can improve the identification rate during automated cutting. First, the input image is preprocessed, during which affine transformation is used to correct the posture of the sugarcane and a rotation matrix is established to obtain the region of interest of the sugarcane. Then, a dataset is built to train the YOLOv3 network model and the position of the stem nodes is initially determined using the YOLOv3 model. Finally, the position of the stem nodes is further located accurately. In this step, a new gradient operator is proposed to extract the edge of the image after YOLOv3 recognition. Then, a local threshold determination method is proposed, which is used to binarize the image after edge extraction. Finally, a localization algorithm for stem nodes is designed to accurately determine the number and location of the stem nodes. The experimental results show that the precision rate, recall rate, and harmonic mean of the stem node recognition algorithm in this paper are 99.68%, 100%, and 99.84%, respectively. Compared to the YOLOv3 network, the precision rate and the harmonic mean are improved by 2.28% and 1.13%, respectively. Compared to other methods introduced in this paper, this algorithm has the highest recognition rate. [ABSTRACT FROM AUTHOR]

Pig counting is an important task in pig sales and breeding supervision. Currently, manual counting is low-efficiency and high-cost and presents challenges in terms of statistical analysis. In response to the difficulties faced in pig part feature detection, the loss of tracking due to rapid movement, and the large counting deviation in pig video tracking and counting research, this paper proposes an improved pig counting algorithm (Mobile Pig Counting Algorithm with YOLOv5xpig and DeepSORTPig (MPC-YD)) based on YOLOv5 + DeepSORT model. The algorithm improves the detection rate of pig body parts by adding two different sizes of SPP networks and using SoftPool instead of MaxPool operations in YOLOv5x. In addition, the algorithm includes a pig reidentification network, a pig-tracking method based on spatial state correction, and a pig counting method based on frame number judgment on the DeepSORT algorithm to improve pig tracking accuracy. Experimental analysis shows that the MPC-YD algorithm achieves an average precision of 99.24% in pig object detection and an accuracy of 85.32% in multitarget pig tracking. In the aisle environment of the slaughterhouse, the MPC-YD algorithm achieves a correlation coefficient (R2) of 98.14% in pig counting from video, and it achieves stable pig counting in a breeding environment. The algorithm has a wide range of application prospects. [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]

The accuracy of pose estimation from feature-based Visual Odometry (VO) algorithms is affected by several factors such as lighting conditions and outliers in the matched features. In this paper, a generic image processing pipeline is proposed to enhance the accuracy and robustness of feature-based VO algorithms. The pipeline consists of three stages, each addressing a problem that affects the performance of VO algorithms. The first stage tackles the lighting condition problem, where a filter called Contrast Limited Adaptive Histogram Equalization (CLAHE) is applied to the images to overcome changes in lighting in the environment. The second stage uses the Suppression via Square Covering (SSC) algorithm to ensure the features are distributed properly over the images. The last stage proposes a novel outliers rejection approach called the Angle-based Outlier Rejection (AOR) algorithm to remove the outliers generated in the feature matching process. The proposed pipeline is generic and modular and can be integrated with any type of feature-based VO (monocular, RGB-D, or stereo). The efficiency of the proposed pipeline is validated using sequences from KITTI (for stereo VO) and TUM (for RGB-D VO) datasets, as well as experimental sequences using an omnidirectional mobile robot (for monocular VO). The obtained results showed the performance gained by enhancing the accuracy and robustness of the VO algorithms without compromising on the computational cost using the proposed pipeline. The results are substantially better as opposed to not using the pipeline. [ABSTRACT FROM AUTHOR]

Human pose estimation refers to the estimation of the location of body parts and how they are connected in an image. Human pose estimation from monocular images has wide applications (e.g., image indexing). Several surveys on human pose estimation can be found in the literature, but they focus on a certain category; for example, model-based approaches or human motion analysis, etc. As far as we know, an overall review of this problem domain has yet to be provided. Furthermore, recent advancements based on deep learning have brought novel algorithms for this problem. In this paper, a comprehensive survey of human pose estimation from monocular images is carried out including milestone works and recent advancements. Based on one standard pipeline for the solution of computer vision problems, this survey splits the problem into several modules: feature extraction and description, human body models, and modeling methods. Problem modeling methods are approached based on two means of categorization in this survey. One way to categorize includes top-down and bottom-up methods, and another way includes generative and discriminative methods. Considering the fact that one direct application of human pose estimation is to provide initialization for automatic video surveillance, there are additional sections for motion-related methods in all modules: motion features, motion models, and motion-based methods. Finally, the paper also collects 26 publicly available data sets for validation and provides error measurement methods that are frequently used. [ABSTRACT FROM AUTHOR]

An essential capability for an unmanned aerial vehicle (UAV) to extend its airborne duration without increasing the size of the aircraft is called the autonomous aerial refueling (AAR). This paper proposes a sensor-in-the-loop, non-tracking method for probe-and-drogue style autonomous aerial refueling tasks by combining sensitivity adjustments of a 3D Flash LIDAR camera with computer vision based image-processing techniques. The method overcomes the inherit ambiguity issues when reconstructing 3D information from traditional 2D images by taking advantage of ready to use 3D point cloud data from the camera, followed by well-established computer vision techniques. These techniques include curve fitting algorithms and outlier removal with the random sample consensus (RANSAC) algorithm to reliably estimate the drogue center in 3D space, as well as to establish the relative position between the probe and the drogue. To demonstrate the feasibility of the proposed method on a real system, a ground navigation robot was designed and fabricated. Results presented in the paper show that using images acquired from a 3D Flash LIDAR camera as real time visual feedback, the ground robot is able to track a moving simulated drogue and continuously narrow the gap between the robot and the target autonomously. [ABSTRACT FROM AUTHOR]

From object tracking to 3D reconstruction, RGB-Depth (RGB-D) camera networks play an increasingly important role in many vision and graphics applications. Practical applications often use sparsely-placed cameras to maximize visibility, while using as few cameras as possible to minimize cost. In general, it is challenging to calibrate sparse camera networks due to the lack of shared scene features across different camera views. In this paper, we propose a novel algorithm that can accurately and rapidly calibrate the geometric relationships across an arbitrary number of RGB-D cameras on a network. Our work has a number of novel features. First, to cope with the wide separation between different cameras, we establish view correspondences by using a spherical calibration object. We show that this approach outperforms other techniques based on planar calibration objects. Second, instead of modeling camera extrinsic calibration using rigid transformation, which is optimal only for pinhole cameras, we systematically test different view transformation functions including rigid transformation, polynomial transformation and manifold regression to determine the most robust mapping that generalizes well to unseen data. Third, we reformulate the celebrated bundle adjustment procedure to minimize the global 3D reprojection error so as to fine-tune the initial estimates. Finally, our scalable client-server architecture is computationally efficient: the calibration of a five-camera system, including data capture, can be done in minutes using only commodity PCs. Our proposed framework is compared with other state-of-the-arts systems using both quantitative measurements and visual alignment results of the merged point clouds. [ABSTRACT FROM AUTHOR]

In this paper, we present a monocular vision-based height estimation algorithm for terrain following flights. The impressive growth of Unmanned Aerial Vehicle (UAV) usage, notably in mapping applications, will soon require the creation of new technologies to enable these systems to better perceive their surroundings. Specifically, we chose to tackle the terrain following problem, as it is still unresolved for consumer available systems. Virtually every mapping aircraft carries a camera; therefore, we chose to exploit this in order to use presently available hardware to extract the height information toward performing terrain following flights. The proposed methodology consists of using optical flow to track features from videos obtained by the UAV, as well as its motion information to estimate the flying height. To determine if the height estimation is reliable, we trained a decision tree that takes the optical flow information as input and classifies whether the output is trustworthy or not. The classifier achieved accuracies of 80% for positives and 90% for negatives, while the height estimation algorithm presented good accuracy. [ABSTRACT FROM AUTHOR]

Today, computer vision algorithms are very important for different fields and applications, such as closed-circuit television security, health status monitoring, and recognizing a specific person or object and robotics. Regarding this topic, the present paper deals with a recent review of the literature on computer vision algorithms (recognition and tracking of faces, bodies, and objects) oriented towards socially assistive robot applications. The performance, frames per second (FPS) processing speed, and hardware implemented to run the algorithms are highlighted by comparing the available solutions. Moreover, this paper provides general information for researchers interested in knowing which vision algorithms are available, enabling them to select the one that is most suitable to include in their robotic system applications. [ABSTRACT FROM AUTHOR]

Recent progress in deep learning has led to accurate and efficient generic object detection networks. Training of highly reliable models depends on large datasets with highly textured and rich images. However, in real-world scenarios, the performance of the generic object detection system decreases when (i) occlusions hide the objects, (ii) objects are present in low-light images, or (iii) they are merged with background information. In this paper, we refer to all these situations as challenging environments. With the recent rapid development in generic object detection algorithms, notable progress has been observed in the field of deep learning-based object detection in challenging environments. However, there is no consolidated reference to cover the state of the art in this domain. To the best of our knowledge, this paper presents the first comprehensive overview, covering recent approaches that have tackled the problem of object detection in challenging environments. Furthermore, we present a quantitative and qualitative performance analysis of these approaches and discuss the currently available challenging datasets. Moreover, this paper investigates the performance of current state-of-the-art generic object detection algorithms by benchmarking results on the three well-known challenging datasets. Finally, we highlight several current shortcomings and outline future directions. [ABSTRACT FROM AUTHOR]

Image-based localization is one of the most widely researched localization techniques in the robotics and computer vision communities. As enormous image data sets are provided through the Internet, many studies on estimating a location with a pre-built image-based 3D map have been conducted. Most research groups use numerous image data sets that contain sufficient features. In contrast, this paper focuses on image-based localization in the case of insufficient images and features. A more accurate localization method is proposed based on a probabilistic map using 3D-to-2D matching correspondences between a map and a query image. The probabilistic feature map is generated in advance by probabilistic modeling of the sensor system as well as the uncertainties of camera poses. Using the conventional PnP algorithm, an initial camera pose is estimated on the probabilistic feature map. The proposed algorithm is optimized from the initial pose by minimizing Mahalanobis distance errors between features from the query image and the map to improve accuracy. To verify that the localization accuracy is improved, the proposed algorithm is compared with the conventional algorithm in a simulation and realenvironments. [ABSTRACT FROM AUTHOR]

Knowledge of crop abiotic and biotic stress is important for optimal irrigation management. While spectral reflectance and infrared thermometry provide a means to quantify crop stress remotely, these measurements can be cumbersome. Computer vision offers an inexpensive way to remotely detect crop stress independent of vegetation cover. This paper presents a technique using computer vision to detect disease stress in wheat. Digital images of differentially stressed wheat were segmented into soil and vegetation pixels using expectation maximization (EM). In the first season, the algorithm to segment vegetation from soil and distinguish between healthy and stressed wheat was developed and tested using digital images taken in the field and later processed on a desktop computer. In the second season, a wireless camera with near real-time computer vision capabilities was tested in conjunction with the conventional camera and desktop computer. For wheat irrigated at different levels and inoculated with wheat streak mosaic virus (WSMV), vegetation hue determined by the EM algorithm showed significant effects from irrigation level and infection. Unstressed wheat had a higher hue (118.32) than stressed wheat (111.34). In the second season, the hue and cover measured by the wireless computer vision sensor showed significant effects from infection (p = 0.0014), as did the conventional camera (p < 0.0001). Vegetation hue obtained through a wireless computer vision system in this study is a viable option for determining biotic crop stress in irrigation scheduling. Such a low-cost system could be suitable for use in the field in automated irrigation scheduling applications. [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]

In low illumination situations, insufficient light in the monitoring device results in poor visibility of effective information, which cannot meet practical applications. To overcome the above problems, a detail preserving low illumination video image enhancement algorithm based on dark channel prior is proposed in this paper. First, a dark channel refinement method is proposed, which is defined by imposing a structure prior to the initial dark channel to improve the image brightness. Second, an anisotropic guided filter (AnisGF) is used to refine the transmission, which preserves the edges of the image. Finally, a detail enhancement algorithm is proposed to avoid the problem of insufficient detail in the initial enhancement image. To avoid video flicker, the next video frames are enhanced based on the brightness of the first enhanced frame. Qualitative and quantitative analysis shows that the proposed algorithm is superior to the contrast algorithm, in which the proposed algorithm ranks first in average gradient, edge intensity, contrast, and patch-based contrast quality index. It can be effectively applied to the enhancement of surveillance video images and for wider computer vision applications. [ABSTRACT FROM AUTHOR]

Event-based vision sensors show great promise for use in embedded applications requiring low-latency passive sensing at a low computational cost. In this paper, we present an event-based algorithm that relies on an Extended Kalman Filter for 6-Degree of Freedom sensor pose estimation. The algorithm updates the sensor pose event-by-event with low latency (worst case of less than 2 μs on an FPGA). Using a single handheld sensor, we test the algorithm on multiple recordings, ranging from a high contrast printed planar scene to a more natural scene consisting of objects viewed from above. The pose is accurately estimated under rapid motions, up to 2.7 m/s. Thereafter, an extension to multiple sensors is described and tested, highlighting the improved performance of such a setup, as well as the integration with an off-the-shelf mapping algorithm to allow point cloud updates with a 3D scene and enhance the potential applications of this visual odometry solution. [ABSTRACT FROM AUTHOR]

Most man-made objects are composed of a few basic geometric primitives (GPs) such as spheres, cylinders, planes, ellipsoids, or cones. Thus, the object recognition problem can be considered as one of geometric primitives extraction. Among the different geometric primitives, cylinders are the most frequently used GPs in real-world scenes. Therefore, cylinder detection and extraction are of great importance in 3D computer vision. Despite the rapid progress of cylinder detection algorithms, there are still two open problems in this area. First, a robust strategy is needed for the initial sample selection component of the cylinder extraction module. Second, detecting multiple cylinders simultaneously has not yet been investigated in depth. In this paper, a robust solution is provided to address these problems. The proposed solution is divided into three sub-modules. The first sub-module is a fast and accurate normal vector estimation algorithm from raw depth images. With the estimation method, a closed-form solution is provided for computing the normal vector at each point. The second sub-module benefits from the maximally stable extremal regions (MSER) feature detector to simultaneously detect cylinders present in the scene. Finally, the detected cylinders are extracted using the proposed cylinder extraction algorithm. Quantitative and qualitative results show that the proposed algorithm outperforms the baseline algorithms in each of the following areas: normal estimation, cylinder detection, and cylinder extraction. [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]

Depth information has been used in computer vision for a wide variety of tasks. Since active range sensors are currently available at low cost, high-quality depth maps can be used as relevant input for many applications. Background subtraction and video segmentation algorithms can be improved by fusing depth and color inputs, which are complementary and allow one to solve many classic color segmentation issues. In this paper, we describe one fusion method to combine color and depth based on an advanced color-based algorithm. This technique has been evaluated by means of a complete dataset recorded with Microsoft Kinect, which enables comparison with the original method. The proposed method outperforms the others in almost every test, showing more robustness to illumination changes, shadows, reflections and camouflage. [ABSTRACT FROM AUTHOR]

In some automatic scene analysis applications, the presence of shadows becomes a nuisance that is necessary to deal with. As a consequence, a preliminary stage in many computer vision algorithms is to attenuate their effect. In this paper, we focus our attention on the detection of shadows cast by static objects outdoors, as the scene is viewed for extended periods of time (days, weeks) from a fixed camera and considering daylight intervals where the main source of light is the sun. In this context, we report two contributions. First, we introduce the use of synthetic images for which ground truth can be generated automatically, avoiding the tedious effort of manual annotation. Secondly, we report a novel application of the intrinsic image concept to the automatic detection of shadows cast by static objects in outdoors. We make both a quantitative and a qualitative evaluation of several algorithms based on this image representation. For the quantitative evaluation, we used the synthetic data set, while for the qualitative evaluation we used both data sets. Our experimental results show that the evaluated methods can partially solve the problem of shadow detection. [ABSTRACT FROM AUTHOR]

This paper focuses on developing a novel technique based on machine vision for detection of foreign substances in injections. Mechanical control yields spin/stop movement of injections which helps to cause relative movement between foreign substances in liquid and an ampoule bottle. Foreign substances are classified into two categories: subsiding-slowly object and subsiding-fast object. A sequence of frames are captured by a camera and used to recognize foreign substances. After image preprocessing like noise reduction and motion detection, two different methods, Moving-object Clustering (MC) and Frame Difference, are proposed to detect the two categories respectively. MC is operated to cluster subsiding-slowly foreign substances, based on the invariant features of those objects. Frame Difference is defined to calculate the difference between two frames due to the change of subsiding-fast objects. 200 ampoule samples filled with injection are tested and the experimental result indicates that the approach can detect the visible foreign substances effectively. [ABSTRACT FROM AUTHOR]

To produce a realistic augmentation in Augmented Reality, the correct relative positions of real objects and virtual objects are very important. In this paper, we propose a novel real-time occlusion handling method based on an object tracking approach. Our method is divided into three steps: selection of the occluding object, object tracking and occlusion handling. The user selects the occluding object using an interactive segmentation method. The contour of the selected object is then tracked in the subsequent frames in realtime. In the occlusion handling step, all the pixels on the tracked object are redrawn on the unprocessed augmented image to produce a new synthesized image in which the relative position between the real and virtual object is correct. The proposed method has several advantages. First, it is robust and stable, since it remains effective when the camera is moved through large changes of viewing angles and volumes or when the object and the background have similar colors. Second, it is fast, since the real object can be tracked in real-time. Last, a smoothing technique provides seamless merging between the augmented and virtual object. Several experiments are provided to validate the performance of the proposed method. [ABSTRACT FROM AUTHOR]

In light of the increasing availability of commercial high-resolution imaging sensors, automatic interpretation tools are needed to extract road features. Currently, many approaches for road extraction are available, but it is acknowledged that there is no single method that would be successful in extracting all types of roads from any remotely sensed imagery. In this paper, a novel classification of roads is proposed, based on both the roads' geometrical, radiometric properties and the characteristics of the sensors. Subsequently, a general road tracking framework is proposed, and one or more suitable road trackers are designed or combined for each type of roads. Extensive experiments are performed to extract roads from aerial/satellite imagery, and the results show that a combination strategy can automatically extract more than 60% of the total roads from very high resolution imagery such as QuickBird and DMC images, with a time-saving of approximately 20%, and acceptable spatial accuracy. It is proven that a combination of multiple algorithms is more reliable, more efficient and more robust for extracting road networks from multiple-source remotely sensed imagery than the individual algorithms. [ABSTRACT FROM AUTHOR]

Monitoring driver attention using the gaze estimation is a typical approach used on road scenes. This indicator is of great importance for safe driving, specially on Level 3 and Level 4 automation systems, where the take over request control strategy could be based on the driver's gaze estimation. Nowadays, gaze estimation techniques used in the state-of-the-art are intrusive and costly, and these two aspects are limiting the usage of these techniques on real vehicles. To test this kind of application, there are some databases focused on critical situations in simulation, but they do not show real accidents because of the complexity and the danger to record them. Within this context, this paper presents a low-cost and non-intrusive camera-based gaze mapping system integrating the open-source state-of-the-art OpenFace 2.0 Toolkit to visualize the driver focalization on a database composed of recorded real traffic scenes through a heat map using NARMAX (Nonlinear AutoRegressive Moving Average model with eXogenous inputs) to establish the correspondence between the OpenFace 2.0 parameters and the screen region the user is looking at. This proposal is an improvement of our previous work, which was based on a linear approximation using a projection matrix. The proposal has been validated using the recent and challenging public database DADA2000, which has 2000 video sequences with annotated driving scenarios based on real accidents. We compare our proposal with our previous one and with an expensive desktop-mounted eye-tracker, obtaining on par results. We proved that this method can be used to record driver attention databases. [ABSTRACT FROM AUTHOR]

With the development of human motion capture (MoCap) equipment and motion analysis technologies, MoCap systems have been widely applied in many fields, including biomedicine, computer vision, virtual reality, etc. With the rapid increase in MoCap data collection in different scenarios and applications, effective segmentation of MoCap data is becoming a crucial issue for further human motion posture and behavior analysis, which requires both robustness and computation efficiency in the algorithm design. In this paper, we propose an unsupervised segmentation algorithm based on limb-bone partition angle body structural representation and autoregressive moving average (ARMA) model fitting. The collected MoCap data were converted into the angle sequence formed by the human limb-bone partition segment and the central spine segment. The limb angle sequences are matched by the ARMA model, and the segmentation points of the limb angle sequences are distinguished by analyzing the good of fitness of the ARMA model. A medial filtering algorithm is proposed to ensemble the segmentation results from individual limb motion sequences. A set of MoCap measurements were also conducted to evaluate the algorithm including typical body motions collected from subjects of different heights, and were labeled by manual segmentation. The proposed algorithm is compared with the principle component analysis (PCA), K-means clustering algorithm (K-means), and back propagation (BP) neural-network-based segmentation algorithms, which shows higher segmentation accuracy due to a more semantic description of human motions by limb-bone partition angles. The results highlight the efficiency and performance of the proposed algorithm, and reveals the potentials of this segmentation model on analyzing inter- and intra-motion sequence distinguishing. [ABSTRACT FROM AUTHOR]

This paper presents an approach of depth image segmentation based on the Evolving Principal Component Clustering (EPCC) method, which exploits data locality in an ordered data stream. The parameters of linear prototypes, which are used to describe different clusters, are estimated in a recursive manner. The main contribution of this work is the extension and application of the EPCC to 3D space for recursive and real-time detection of flat connected surfaces based on linear segments, which are all detected in an evolving way. To obtain optimal results when processing homogeneous surfaces, we introduced two-step filtering for outlier detection within a clustering framework and considered the noise model, which allowed for the compensation of characteristic uncertainties that are introduced into the measurements of depth sensors. The developed algorithm was compared with well-known methods for point cloud segmentation. The proposed approach achieves better segmentation results over longer distances for which the signal-to-noise ratio is low, without prior filtering of the data. On the given database, an average rate higher than 90% was obtained for successfully detected flat surfaces, which indicates high performance when processing huge point clouds in a non-iterative manner. [ABSTRACT FROM AUTHOR]

The automation strategy of today's smart cities relies on large IoT (internet of Things) systems that collect big data analytics to gain insights. Although there have been recent reviews in this field, there is a remarkable gap that addresses four sides of the problem. Namely, the application of video surveillance in smart cities, algorithms, datasets, and embedded systems. In this paper, we discuss the latest datasets used, the algorithms used, and the recent advances in embedded systems to form edge vision computing are introduced. Moreover, future trends and challenges are addressed. [ABSTRACT FROM AUTHOR]

Technologies and services towards smart-vehicles and Intelligent-Transportation-Systems (ITS), continues to revolutionize many aspects of human life. This paper presents a detailed survey of current techniques and advancements in Automatic-Number-Plate-Recognition (ANPR) systems, with a comprehensive performance comparison of various real-time tested and simulated algorithms, including those involving computer vision (CV). ANPR technology has the ability to detect and recognize vehicles by their number-plates using recognition techniques. Even with the best algorithms, a successful ANPR system deployment may require additional hardware to maximize its accuracy. The number plate condition, non-standardized formats, complex scenes, camera quality, camera mount position, tolerance to distortion, motion-blur, contrast problems, reflections, processing and memory limitations, environmental conditions, indoor/outdoor or day/night shots, software-tools or other hardware-based constraint may undermine its performance. This inconsistency, challenging environments and other complexities make ANPR an interesting field for researchers. The Internet-of-Things is beginning to shape future of many industries and is paving new ways for ITS. ANPR can be well utilized by integrating with RFID-systems, GPS, Android platforms and other similar technologies. Deep-Learning techniques are widely utilized in CV field for better detection rates. This research aims to advance the state-of-knowledge in ITS (ANPR) built on CV algorithms; by citing relevant prior work, analyzing and presenting a survey of extraction, segmentation and recognition techniques whilst providing guidelines on future trends in this area. [ABSTRACT FROM AUTHOR]

Due to deep learning's accurate cognition of the street environment, the convolutional neural network has achieved dramatic development in the application of street scenes. Considering the needs of autonomous driving and assisted driving, in a general way, computer vision technology is used to find obstacles to avoid collisions, which has made semantic segmentation a research priority in recent years. However, semantic segmentation has been constantly facing new challenges for quite a long time. Complex network depth information, large datasets, real-time requirements, etc., are typical problems that need to be solved urgently in the realization of autonomous driving technology. In order to address these problems, we propose an improved lightweight real-time semantic segmentation network, which is based on an efficient image cascading network (ICNet) architecture, using multi-scale branches and a cascaded feature fusion unit to extract rich multi-level features. In this paper, a spatial information network is designed to transmit more prior knowledge of spatial location and edge information. During the course of the training phase, we append an external loss function to enhance the learning process of the deep learning network system as well. This lightweight network can quickly perceive obstacles and detect roads in the drivable area from images to satisfy autonomous driving characteristics. The proposed model shows substantial performance on the Cityscapes dataset. With the premise of ensuring real-time performance, several sets of experimental comparisons illustrate that SP-ICNet enhances the accuracy of road obstacle detection and provides nearly ideal prediction outputs. Compared to the current popular semantic segmentation network, this study also demonstrates the effectiveness of our lightweight network for road obstacle detection in autonomous driving. [ABSTRACT FROM AUTHOR]

Rigid registration of 3D point clouds is the key technology in robotics and computer vision. Most commonly, the iterative closest point (ICP) and its variants are employed for this task. These methods assume that the closest point is the corresponding point and lead to sensitivity to the outlier and initial pose, while they have poor computational efficiency due to the closest point computation. Most implementations of the ICP algorithm attempt to deal with this issue by modifying correspondence or adding coarse registration. However, this leads to sacrificing the accuracy rate or adding the algorithm complexity. This paper proposes a hierarchical optimization approach that includes improved voxel filter and Multi-Scale Voxelized Generalized-ICP (MVGICP) for 3D point cloud registration. By combining traditional voxel sampling with point density, the outlier filtering and downsample are successfully realized. Through multi-scale iteration and avoiding closest point computation, MVGICP solves the local minimum problem and optimizes the operation efficiency. The experimental results demonstrate that the proposed algorithm is superior to the current algorithms in terms of outlier filtering and registration performance. [ABSTRACT FROM AUTHOR]

The embedded visual tracking system has higher requirements for real-time performance and system resources, and this is a challenge for visual tracking systems with available hardware resources. The major focus of this study is evaluating the results of hardware optimization methods. These optimization techniques provide efficient utilization based on limited hardware resources. This paper also uses a pragmatic approach to investigate the real-time performance effect by implementing and optimizing a kernel correlation filter (KCF) tracking algorithm based on a vision digital signal processor (vision DSP). We examine and analyze the impact factors of the tracking system, which include DP (data parallelism), IP (instruction parallelism), and the characteristics of parallel processing of the DSP core and iDMA (integrated direct memory access). Moreover, we utilize a time-sharing strategy to increase the system runtime speed. These research results are also applicable to other machine vision algorithms. In addition, we introduced a scale filter to overcome the disadvantages of KCF for scale transformation. The experimental results demonstrate that the use of system resources and real-time tracking speed also satisfies the expected requirements, and the tracking algorithm with a scale filter can realize almost the same accuracy as the DSST (discriminative scale space tracking) algorithm under a vision DSP environment. [ABSTRACT FROM AUTHOR]

In recent years, deep learning models have achieved remarkable successes in various applications, such as pattern recognition, computer vision, and signal processing. However, high-performance deep architectures are often accompanied by a large storage space and long computational time, which make it difficult to fully exploit many deep neural networks (DNNs), especially in scenarios in which computing resources are limited. In this paper, to tackle this problem, we introduce a method for compressing the structure and parameters of DNNs based on neuron agglomerative clustering (NAC). Specifically, we utilize the agglomerative clustering algorithm to find similar neurons, while these similar neurons and the connections linked to them are then agglomerated together. Using NAC, the number of parameters and the storage space of DNNs are greatly reduced, without the support of an extra library or hardware. Extensive experiments demonstrate that NAC is very effective for the neuron agglomeration of both the fully connected and convolutional layers, which are common building blocks of DNNs, delivering similar or even higher network accuracy. Specifically, on the benchmark CIFAR-10 and CIFAR-100 datasets, using NAC to compress the parameters of the original VGGNet by 92.96% and 81.10%, respectively, the compact network obtained still outperforms the original networks. [ABSTRACT FROM AUTHOR]

Current research on dynamic displacement measurement based on computer vision mostly requires professional high-speed cameras and an ideal shooting environment to ensure the performance and accuracy of the analysis. However, the high cost of the camera and strict requirements of sharp image contrast and stable environment during the shooting process limit the broad application of the technology. This paper proposes an improved vision method to implement multi-point dynamic displacement measurements with smartphones in an interference environment. A motion-enhanced spatio-temporal context (MSTC) algorithm is developed and applied together with the optical flow (OF) algorithm to realize a simultaneous tracking and dynamic displacement extraction of multiple points on a vibrating structure in the interference environment. Finally, a sine-sweep vibration experiment on a cantilever sphere model is presented to validate the feasibility of the proposed method in a wide-band frequency range. In the test, a smartphone was used to shoot the vibration process of the sine-sweep-excited sphere, and illumination change, fog interference, and camera jitter were artificially simulated to represent the interference environment. The results of the proposed method are compared to conventional displacement sensor data and current vision method results. It is demonstrated that, in an interference environment, (1) the OF method is prone to mismatch the feature points and leads to data deviated or lost; (2) the conventional STC method is sensitive to target selection and can effectively track those targets having a large proportion of pixels in the context with motion tendency similar to the target center; (3) the proposed MSTC method, however, can ease the sensitivity to target selection through in-depth processing of the information in the context and finally enhance the robustness of the target tracking. In addition, the MSTC method takes less than one second to track each target between adjacent frame images, implying a potential for online measurement. [ABSTRACT FROM AUTHOR]

In recent years, machine vision algorithms have played an influential role as core technologies in several practical applications, such as surveillance, autonomous driving, and object recognition/localization. However, as almost all such algorithms are applicable to clear weather conditions, their performance is severely affected by any atmospheric turbidity. Several image visibility restoration algorithms have been proposed to address this issue, and they have proven to be a highly efficient solution. This paper proposes a novel method to recover clear images from degraded ones. To this end, the proposed algorithm uses a supervised machine learning-based technique to estimate the pixel-wise extinction coefficients of the transmission medium and a novel compensation scheme to rectify the post-dehazing false enlargement of white objects. Also, a corresponding hardware accelerator implemented on a Field Programmable Gate Array chip is in order for facilitating real-time processing, a critical requirement of practical camera-based systems. Experimental results on both synthetic and real image datasets verified the proposed method's superiority over existing benchmark approaches. Furthermore, the hardware synthesis results revealed that the accelerator exhibits a processing rate of nearly 271.67 Mpixel/s, enabling it to process 4K videos at 30.7 frames per second in real time. [ABSTRACT FROM AUTHOR]

Most recent computer vision tasks take into account the distribution of image features to obtain more powerful models and better performance. One of the most commonly used techniques to this purpose is the diffusion algorithm, which fuses manifold data and k-Nearest Neighbors (kNN) graphs. In this paper, we describe how we optimized diffusion in an image retrieval task aimed at mobile vision applications, in order to obtain a good trade-off between computation load and performance. From a computational efficiency viewpoint, the high complexity of the exhaustive creation of a full kNN graph for a large database renders such a process unfeasible on mobile devices. From a retrieval performance viewpoint, the diffusion parameters are strongly task-dependent and affect significantly the algorithm performance. In the method we describe herein, we tackle the first issue by using approximate algorithms in building the kNN tree. The main contribution of this work is the optimization of diffusion parameters using a genetic algorithm (GA), which allows us to guarantee high retrieval performance in spite of such a simplification. The results we have obtained confirm that the global search for the optimal diffusion parameters performed by a genetic algorithm is equivalent to a massive analysis of the diffusion parameter space for which an exhaustive search would be totally unfeasible. We show that even a grid search could often be less efficient (and effective) than the GA, i.e., that the genetic algorithm most often produces better diffusion settings when equal computing resources are available to the two approaches. Our method has been tested on several publicly-available datasets: Oxford5k, R Oxford5k, Paris6k, R Paris6k, and Oxford105k, and compared to other mainstream approaches. [ABSTRACT FROM AUTHOR]

As the development of object detection technology in computer vision, identifying objects is always an active yet challenging task, and even more efficient and accurate requirements are being imposed on state-of-the-art algorithms. However, many algorithms perform object box regression based on RPN(Region Proposal Network) and anchors, which cannot accurately describe the shape information of the object. In this paper, we propose a new object detection method called Field Network (FN) and Region Fitting Algorithm (RFA). It can solve these problems by Center Field. Center field reflects the probability of the pixel approaching the object center. Different from the previous methods, we abandoned anchors and ROI technologies, and propose the concept of Field. Field is the intensity of the object area, reflecting the probability of the object in the area. Based on the distribution of the probability density of the object center in the visual field perception area, we add the Object Field in the output part. And we abstract it into an Elliptic Field with normal distribution and use RFA to fit objects. Additionally, we add two fields to predict the x,y components of the object direction which contain the neural units in the field array. We extract the objects through these Fields. Moreover, our model is relatively simple and have smaller size, which is only 73 M. Our method improves performance considerably over baseline systems on DOTA, MS COCO and PASCAL VOC datasets, with overall performance competitive with recent state-of-the-art systems. [ABSTRACT FROM AUTHOR]

For a safe market launch of automated vehicles, the risks of the overall system as well as the sub-components must be efficiently identified and evaluated. This also includes camera-based object detection using artificial intelligence algorithms. It is trivial and explainable that due to the principle of the camera, performance depends highly on the environmental conditions and can be poor, for example in heavy fog. However, there are other factors influencing the performance of camera-based object detection, which will be comprehensively investigated for the first time in this paper. Furthermore, a precise modeling of the detection performance and the explanation of individual detection results is not possible due to the artificial intelligence based algorithms used. Therefore, a modeling approach based on the investigated influence factors is proposed and the newly developed SHapley Additive exPlanations (SHAP) approach is adopted to analyze and explain the detection performance of different object detection algorithms. The results show that many influence factors such as the relative rotation of an object towards the camera or the position of an object on the image have basically the same influence on the detection performance regardless of the detection algorithm used. In particular, the revealed weaknesses of the tested object detectors can be used to derive challenging and critical scenarios for the testing and type approval of automated vehicles. [ABSTRACT FROM AUTHOR]

Microsoft Kinect, a low-cost motion capture device, has huge potential in applications that require machine vision, such as human-robot interactions, home-based rehabilitation and clinical assessments. The Kinect sensor can track 25 key three-dimensional (3D) "skeleton" joints on the human body at 30 frames per second, and the skeleton data often have acceptable accuracy. However, the skeleton data obtained from the sensor sometimes exhibit a high level of jitter due to noise and estimation error. This jitter is worse when there is occlusion or a subject moves slightly out of the field of view of the sensor for a short period of time. Therefore, this paper proposed a novel approach to simultaneously handle the noise and error in the skeleton data derived from Kinect. Initially, we adopted classification processing to divide the skeleton data into noise data and erroneous data. Furthermore, we used a Kalman filter to smooth the noise data and correct erroneous data. We performed an occlusion experiment to prove the effectiveness of our algorithm. The proposed method outperforms existing techniques, such as the moving mean filter and traditional Kalman filter. The experimental results show an improvement of accuracy of at least 58.7%, 47.5% and 22.5% compared to the original Kinect data, moving mean filter and traditional Kalman filter, respectively. Our method provides a new perspective for Kinect data processing and a solid data foundation for subsequent research that utilizes Kinect. [ABSTRACT FROM AUTHOR]

Vibrations often cause visual fatigue for drivers, and measuring the relative motion between the driver and the display is important for evaluating this visual fatigue. This paper proposes a non-contact videometric measurement method for studying the three-dimensional trajectories of the driver's eyes based on stereo vision. The feasibility of this method is demonstrated by dynamic calibration. A high-speed dual-camera image acquisition system is used to obtain high-definition images of the face, and the relative trajectories between the eyes and the display are obtained by a set of robust algorithms. The trajectories of the eyes in three-dimensional space are then reconstructed during the vehicle driving process. This new approach provides three-dimensional information and is effective for assessing how vibration affects human visual performance. [ABSTRACT FROM AUTHOR]

The research field of visual-inertial odometry has entered a mature stage in recent years. However, unneglectable problems still exist. Tradeoffs have to be made between high accuracy and low computation for users. In addition, notation confusion exists in quaternion descriptions of rotation; although not fatal, this may results in unnecessary difficulties in understanding for researchers. In this paper, we develop a visual-inertial odometry which gives consideration to both precision and computation. The proposed algorithm is a filter-based solution that utilizes the framework of the noted multi-state constraint Kalman filter. To dispel notation confusion, we deduced the error state transition equation from scratch, using the more cognitive Hamilton notation of quaternion. We further come up with a fully linear closed-form formulation that is readily implemented. As the filter-based back-end is vulnerable to feature matching outliers, a descriptor-assisted optical flow tracking front-end was developed to cope with the issue. This modification only requires negligible additional computation. In addition, an initialization procedure is implemented, which automatically selects static data to initialize the filter state. Evaluations of proposed methods were done on a public, real-world dataset, and comparisons were made with state-of-the-art solutions. The experimental results show that the proposed solution is comparable in precision and demonstrates higher computation efficiency compared to the state-of-the-art. [ABSTRACT FROM AUTHOR]