Search

Automated crowd counting is a crucial aspect of surveillance, especially in the context of mass events attended by large populations. Traditional methods of manually counting the people attending an event are error-prone, necessitating the development of automated methods. Accurately estimating crowd counts across diverse scenes is challenging due to high variations in the sizes of human heads. Regression-based crowd-counting methods often overestimate counts in low-density situations, while detection-based models struggle in high-density scenarios to precisely detect the head. In this work, we propose a unified framework that integrates regression and detection models to estimate the crowd count in diverse scenes. Our approach leverages a routing strategy based on crowd density variations within an image. By classifying image patches into density levels and employing a Patch-Routing Module (PRM) for routing, the framework directs patches to either the Detection or Regression Network to estimate the crowd count. The proposed framework demonstrates superior performance across various datasets, showcasing its effectiveness in handling diverse scenes. By effectively integrating regression and detection models, our approach offers a comprehensive solution for accurate crowd counting in scenarios ranging from low-density to high-density situations.

Indoor scene classification plays a pivotal role in enabling social robots to seamlessly adapt to their environments, facilitating effective navigation and interaction within diverse indoor scenes. By accurately characterizing indoor scenes, robots can autonomously tailor their behaviors, making informed decisions to accomplish specific tasks. Traditional methods relying on manually crafted features encounter difficulties when characterizing complex indoor scenes. On the other hand, deep learning models address the shortcomings of traditional methods by autonomously learning hierarchical features from raw images. Despite the success of deep learning models, existing models still struggle to effectively characterize complex indoor scenes. This is because there is high degree of intra-class variability and inter-class similarity within indoor environments. To address this problem, we propose a dual-stream framework that harnesses both global contextual information and local features for enhanced recognition. The global stream captures high-level features and relationships across the scene. The local stream employs a fully convolutional network to extract fine-grained local information. The proposed dual-stream architecture effectively distinguishes scenes that share similar global contexts but contain different localized objects. We evaluate the performance of the proposed framework on a publicly available benchmark indoor scene dataset. From the experimental results, we demonstrate the effectiveness of the proposed framework.

Whether the computer is driving your car or you are, advanced driver assistance systems (ADAS) come into play on all levels, from weather monitoring to safety. These modern-day ADASs use various assisting tools for drivers to keep the journey safe; these sophisticated tools provide early signals of numerous events, such as road conditions, emerging traffic scenarios, and weather warnings. Many urban applications, such as car-sharing and logistics, rely on accurate and up-to-date road map data. Map generation methods use a variety of data sources, including but not limited to global positioning systems (GPS). In this research we propose a GPS-only data trajectory analysis and a novel scheme to convert GPS trajectory data to image-based data to train a custom Convolutional Neural Network (CNN) model. The empirical results with an extensive 5-fold cross-validation show that the proposed scheme identifies turn and not turn with more than 94% recall. It outperforms the existing turn detection schemes on two major frontiers, the required data and the accuracy achieved in detecting different driving behaviors.

Congestion in public places is one of the major problems in public transportation systems and causes a high level of discomfort for the commuters. Traditionally, overcrowding is detected by manually monitoring and analyzing the video streams from the surveillance cameras, which might lead to errors due to limited human activity. On the other hand, current machine learning models for automatic congestion detection require a massive amount of labeled data to train the network. These models suffer from the over-fitting problem and cannot be generalized to novel scenes. First, we propose a novel synthetic dataset for congestion detection in public places to address these problems. Secondly, we propose a Bidirectional Long-short-term-memory (Bi-LSTM) model that exploits synthetic datasets to boost the performance of congestion detection in the wild. We adopt a domain adaptation strategy to bridge the gap between the real and synthetic data by pre-train the model on the synthetic dataset and then fine-tuning the model on real data. From experiment results, we observe that the proposed framework achieves a significant performance boost on the real datasets after training on the synthetic dataset.

The rapid expansion of the world’s population has resulted in an increased demand for agricultural products which necessitates the need to improve crop yields. To enhance crop yields, it is imperative to control weeds. Traditionally, weed control predominantly relied on the use of herbicides; however, the indiscriminate application of herbicides presents potential hazards to both crop health and productivity. Fortunately, the advent of cutting-edge technologies such as unmanned vehicle technology (UAVs) and computer vision has provided automated and efficient solutions for weed control. These approaches leverage drone images to detect and identify weeds with a certain level of accuracy. Nevertheless, the identification of weeds in drone images poses significant challenges attributed to factors like occlusion, variations in color and texture, and disparities in scale. The utilization of traditional image processing techniques and deep learning approaches, which are commonly employed in existing methods, presents difficulties in extracting features and addressing scale variations. In order to address these challenges, an innovative deep learning framework is introduced which is designed to classify every pixel in a drone image into categories such as weed, crop, and others. In general, our proposed network adopts an encoder–decoder structure. The encoder component of the network effectively combines the Dense-inception network with the Atrous spatial pyramid pooling module, enabling the extraction of multi-scale features and capturing local and global contextual information seamlessly. The decoder component of the network incorporates deconvolution layers and attention units, namely, channel and spatial attention units (CnSAUs), which contribute to the restoration of spatial information and enhance the precise localization of weeds and crops in the images. The performance of the proposed framework is assessed using a publicly available benchmark dataset known for its complexity. The effectiveness of the proposed framework is demonstrated via comprehensive experiments, showcasing its superiority by achieving a 0.81 mean Intersection over Union (mIoU) on the challenging dataset.

Nowadays, digital content like videos, audio and images are widely used as evidence in criminal courts and forensic laboratories. Due to the advanced low-cost and easily available multimedia/communication tools and softwares, manipulation of the content is a no-brain task. Thus, the protection of digital content originality is a challenge for the content owners and researchers before it can be produced in court or used for some other purpose. We proposed a motion vector watermarking technique that validate and authenticate videos. We are embedding the correlated watermark in the integer wavelet transform domain. In our method, the selection of embedding areas is based on the variation of motion vectors. The video frames are fully protected in both spatial and transform domains since the watermark is correlated with the approximation subbands of wavelet transform before embedding. The proposed technique can concisely determine the attacked regions. The results validate the performance of the proposed approach in terms of quality metrics like peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), normalized coefficients (NC) and bit error rate (BER).

Land scene classification in satellite imagery has a wide range of applications in remote surveillance, environment monitoring, remote scene analysis, Earth observations and urban planning. Due to immense advantages of the land scene classification task, several methods have been proposed during recent years to automatically classify land scenes in remote sensing images. Most of the work focuses on designing and developing deep networks to identify land scenes from high-resolution satellite images. However, these methods face challenges in identifying different land scenes. Complex texture, cluttered background, extremely small size of objects and large variations in object scale are the common challenges that restrict the models to achieve high performance. To tackle these challenges, we propose a multi-branch deep learning framework that efficiently combines global contextual features with multi-scale features to identify complex land scenes. Generally, the framework consists of two branches. The first branch extracts global contextual information from different regions of the input image, and the second branch exploits a fully convolutional network (FCN) to extract multi-scale local features. The performance of the proposed framework is evaluated on three benchmark datasets, UC-Merced, SIRI-WHU, and EuroSAT. From the experiments, we demonstrate that the framework achieves superior performance compared to other similar models.

Floods are the most frequent natural disasters, occurring almost every year around the globe. To mitigate the damage caused by a flood, it is important to timely assess the magnitude of the damage and efficiently conduct rescue operations, deploy security personnel and allocate resources to the affected areas. To efficiently respond to the natural disaster, it is very crucial to swiftly obtain accurate information, which is hard to obtain during a post-flood crisis. Generally, high resolution satellite images are predominantly used to obtain post-disaster information. Recently, deep learning models have achieved superior performance in extracting high-level semantic information from satellite images. However, due to the loss of multi-scale and global contextual features, existing deep learning models still face challenges in extracting complete and uninterrupted results. In this work, we proposed a novel deep learning semantic segmentation model that reduces the loss of multi-scale features and enhances global context awareness. Generally, the proposed framework consists of three modules, encoder, decoder and bridge, combined in a popular U-shaped scheme. The encoder and decoder modules of the framework introduce Res-inception units to obtain reliable multi-scale features and employ a bridge module (between the encoder and decoder) to capture global context. To demonstrate the effectiveness of the proposed framework, we perform an evaluation using a publicly available challenging dataset, FloodNet. Furthermore, we compare the performance of the proposed framework with other reference methods. We compare the proposed framework with recent reference models. Quantitative and qualitative results show that the proposed framework outperforms other reference models by an obvious margin.

Seeking high power conversion efficiency is always a striving pathway for many researchers to commercialize organic solar cells (OSCs). Thus, it is essential to achieve high conversion efficiency to make solar oriented devices as truly green energy resource. In this paper, we report novel tandem structures based on organic materials in which homo and hybrid schemes are utilized. The homo tandem structures involve the well-known organic materials such as P3HT: PCBM and PTB7: PCBM as top and bottom absorber layers, while the hybrid tandem structure is made by the combination of P3HT: PCBM and PTB7: PCBM, respectively. Due to different absorption coefficients, simulation results indicate that the hybrid tandem structure exhibit large values for current density Jsc = 22.60mA/cm2, open circuit voltage Voc = 0.85V, fill factor FF = 68.21%, and efficiency η = 13.96%, compared to homo tandem cell and proves to be more efficient and highly stable towards high temperatures, which indicates that the proposed structure is more suitable for practical applications.

Pedestrian head detection plays an important role in identifying and localizing individuals in real world visual data. Head detection is a nontrivial problem due to considerable variance in camera view-points, scales, human poses, and appearances in the scene. Thanks to the translation invariance property of convolutional neural networks (CNNs) which enables large capacity CNNs to handle the problem of appearance and pose variations in the scene. However, the problem of scale invariance is still an open issue. To address this problem, this paper presents a two-stage head detection framework that utilizes fully convolutional network (FCN) to generate scale-aware proposals followed by CNN that classifies each proposal into two classes, i.e. head and background. Experiments results show that using scale-aware proposals obtained by FCN, the object recall rate and mean average precision (mAP) are improved. Additionaly, we demonstrate that our framework achieved state-of-the-art results on four challenging benchmark datasets, i.e. HollywoodHeads, Casablanca, SHOCK, and WIDERFACE.

Crowd counting and density estimation is an important and challenging problem in the visual analysis of the crowd. Most of the existing approaches use regression on density maps for the crowd count from a single image. However, these methods cannot localize individual pedestrian and therefore cannot estimate the actual distribution of pedestrians in the environment. On the other hand, detection-based methods detect and localize pedestrians in the scene, but the performance of these methods degrades when applied in high-density situations. To overcome the limitations of pedestrian detectors, we proposed a motion-guided filter (MGF) that exploits spatial and temporal information between consecutive frames of the video to recover missed detections. Our framework is based on the deep convolution neural network (DCNN) for crowd counting in the low-to-medium density videos. We employ various state-of-the-art network architectures, namely, Visual Geometry Group (VGG16), Zeiler and Fergus (ZF), and VGGM in the framework of a region-based DCNN for detecting pedestrians. After pedestrian detection, the proposed motion guided filter is employed. We evaluate the performance of our approach on three publicly available datasets. The experimental results demonstrate the effectiveness of our approach, which significantly improves the performance of the state-of-the-art detectors.

Counting and localization of people in videos consisting of low density to high density crowds encounter many key challenges including complex backgrounds, scale variations, nonuniform distributions, and occlusions. For this purpose, we propose a scale driven convolutional neural network (SD-CNN) model, which is based on the assumption that heads are the dominant and visible features regardless of the density of crowds. To deal with the problem of different scales of heads in different regions of the videos, we annotate a set of heads in random locations of the videos to develop a scale map representing the mapping of head sizes. We then extract scale aware proposals based on the scale map which are fed to the SD-CNN model acting as a head detector. Our model provides a response matrix rendering accurate head positions via nonmaximal suppression. For experimental evaluations, we consider three standard datasets presenting low density to high density crowd scenes. Our proposed SD-CNN model outperforms the state-of-the-art methods in terms of both frame-level and pixel-level analyses.

Among various physiological signal acquisition methods for the study of the human brain, EEG (Electroencephalography) is more effective. EEG provides a convenient, non-intrusive, and accurate way of capturing brain signals in multiple channels at fine temporal resolution. We propose an ensemble learning algorithm for automatically computing the most discriminative subset of EEG channels for internal emotion recognition. Our method describes an EEG channel using kernel-based representations computed from the training EEG recordings. For ensemble learning, we formulate a graph embedding linear discriminant objective function using the kernel representations. The objective function is efficiently solved via sparse non-negative principal component analysis and the final classifier is learned using the sparse projection coefficients. Our algorithm is useful in reducing the amount of data while improving computational efficiency and classification accuracy at the same time. The experiments on publicly available EEG dataset demonstrate the superiority of the proposed algorithm over the compared methods.

With the evolution of the convolutional neural network (CNN), object detection in the underwater environment has gained a lot of attention. However, due to the complex nature of the underwater environment, generic CNN-based object detectors still face challenges in underwater object detection. These challenges include image blurring, texture distortion, color shift, and scale variation, which result in low precision and recall rates. To tackle this challenge, we propose a detection refinement algorithm based on spatial–temporal analysis to improve the performance of generic detectors by suppressing the false positives and recovering the missed detections in underwater videos. In the proposed work, we use state-of-the-art deep neural networks such as Inception, ResNet50, and ResNet101 to automatically classify and detect the Norway lobster Nephrops norvegicus burrows from underwater videos. Nephrops is one of the most important commercial species in Northeast Atlantic waters, and it lives in burrow systems that it builds itself on muddy bottoms. To evaluate the performance of proposed framework, we collected the data from the Gulf of Cadiz. From experiment results, we demonstrate that the proposed framework effectively suppresses false positives and recovers missed detections obtained from generic detectors. The mean average precision (mAP) gained a 10% increase with the proposed refinement technique.

Person detection in real videos and images is a classical research problem in computer vision. Person detection is a nontrivial problem that offers many challenges due to several nuisances that commonly observed in natural videos. Among these, scale is the main challenging problem in various object detection tasks. To solve the scale problem, we propose a framework that estimates the scales of person’s heads, as we argue that head is the only visible part in complex scenes. we propose a head detection framework that explicitly handles head scales. The framework consists of two sequential networks: (1) scale estimation network (SENet) and (2) head detection network. SENet predicts the distribution of scales from the input image in the form of histogram. Then the scale histogram adjust anchor scale set of region proposal network that generates object proposals. These objects proposals are then classified into two classes, that is, head and background by the detection network. We evaluate proposed framework on three challenging benchmark datasets. Experiment results show that proposed framework achieves state-of-the-art performance.

Land cover semantic segmentation in high-spatial resolution satellite images plays a vital role in efficient management of land resources, smart agriculture, yield estimation and urban planning. With the recent advancement in remote sensing technologies, such as satellites, drones, UAVs, and airborne vehicles, a large number of high-resolution satellite images are readily available. However, these high-resolution satellite images are complex due to increased spatial resolution and data disruption caused by different factors involved in the acquisition process. Due to these challenges, an efficient land-cover semantic segmentation model is difficult to design and develop. In this paper, we develop a hybrid deep learning model that combines the benefits of two deep models, i.e., DenseNet and U-Net. This is carried out to obtain a pixel-wise classification of land cover. The contraction path of U-Net is replaced with DenseNet to extract features of multiple scales, while long-range connections of U-Net concatenate encoder and decoder paths are used to preserve low-level features. We evaluate the proposed hybrid network on a challenging, publicly available benchmark dataset. From the experimental results, we demonstrate that the proposed hybrid network exhibits a state-of-the-art performance and beats other existing models by a considerable margin.

COVID-19 caused the largest economic recession in the history by placing more than one third of world’s population in lockdown. The prolonged restrictions on economic and business activities caused huge economic turmoil that significantly affected the financial markets. To ease the growing pressure on the economy, scientists proposed intermittent lockdowns commonly known as “smart lockdowns”. Under smart lockdown, areas that contain infected clusters of population, namely hotspots, are placed on lockdown, while economic activities are allowed to operate in un-infected areas. In this study, we proposed a novel deep learning prediction framework for the accurate prediction of hotpots. We exploit the benefits of two deep learning models, i.e., Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) and propose a hybrid framework that has the ability to extract multi time-scale features from convolutional layers of CNN. The multi time-scale features are then concatenated and provide as input to 2-layers LSTM model. The LSTM model identifies short, medium and long-term dependencies by learning the representation of time-series data. We perform a series of experiments and compare the proposed framework with other state-of-the-art statistical and machine learning based prediction models. From the experimental results, we demonstrate that the proposed framework beats other existing methods with a clear margin.