Human Behavior Recognition Based on Motion Decomposition

To improve traditional video surveillance systems' performance on human behavior recognition, a new system has been built. Not visual camera but depth camera is chosen as sensor. To adapt to the most common forward oblique view of camera, a normalized digital elevation map, whose pixel intensity indicates the elevation of the scene, is built from the depth image. Coordinates and intensity in the digital elevation map represent the position information about the scene. Oriented templates are proposed to match and detect the human head robustly in the elevation map. As for the excellent visibility of human head in the elevation map, we track the human head to get trajectory. By combining the trajectory information of human head with the elevation map, several predefined human behaviors are recognized. Our behavior recognition method is straightforward and robust. This uniqueness has no similar with the traditional machine learning and classification framework about human behavior recognition.

The Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP) is India's premier conference in Computer Vision, Graphics, Image Processing and related fields. Started in 1998, it is a biennial international conference providing a forum for the presentation of technological advances and research findings in these areas.

Big data analysis of human behavior can provide the basis and support for the application of various scenarios. Using sensors for human behavior analysis is an effective means of identification method, which is very valuable for research. To address the problems of low recognition accuracy, low recognition efficiency of traditional human behavior recognition (HBR) algorithms in complex scenes, in this paper, we propose an HBR algorithm for Mobile Big data analytics in wireless sensor network using improved transfer learning. First, different wireless sensors are fused to obtain human behavior mobile big data, and then by analyzing the importance of human behavior features (HBF), the dynamic change parameters of HBF extraction threshold are calculated. Second, combined with the dynamic change parameters of threshold, the HBF of complex scenes are extracted. Finally, the best classification function of human behavior in complex scenes is obtained by using the classification function of HBF in complex scenes. Human behavior in complex scenes is classified according to the HBF in the feature set. The HBR algorithm is designed by using the improved transfer learning network to realize the recognition of human behavior in complex scenes. The results show that the proposed algorithm can accurately recognize up to 22 HBF points, and can control the HBR time within 2 s. The human behavior false recognition rate of miscellaneous scenes is less than 10%. The recognition speed is above 10/s, and the recall rate can reach more than 98%, which improves the HBR ability of complex scenes.

Background subtraction is important process in many automatic video content analysis applications.extracting foreground moving objects from video sequences is a crucial step and also a hot topic in computer vision and image processing .This paper presents a new approach in image processing for detection crack in video.This method involves two main steps: First, we use based on Visual Background Extractor (Vibe) to detection moving crack in video.Second steps: Using a suitable threshold in a binary image and classifies all pixels two groups' background and foreground.And use filter area and changes the area if less than the specific number to back; and using filtering to elimination of residual noise.This paper describes a method for detection crack in video we use Image Processing Techniques.The advantage of this method is clearly and accurate detection of cracks in video.Experimental work shows that our method is improved relatively to the other widely used techniques.

Falling is a common but fatal human behavior in life. With the rapid growth of the aging population, fall-related human behavior recognition has been extensively investigated using radar. Nevertheless, human behavior recognition frequently exhibits suboptimal generalization capabilities due to the scarcity of labeled data. To address this problem, we propose physics-informed data augmentation (PIDA) for the micro-Doppler signature (m-DS) produced from frequency-modulated continuous wave (FMCW) radar to facilitate domain generalization (DG) of human behavior recognition. Specifically, leveraging classical electromagnetic wave propagation theory and radar signal processing principles, we introduce a PIDA strategy: 1) Distance Data Augmentation (DDA), which simulates electromagnetic attenuation effects across varying detection ranges, and 2) Behavior Pattern Data Augmentation (BPDA), designed to preserve intrinsic kinematic features while diversifying motion styles. Subsequently, combined with the proposed PIDA, a multi-source domain adversarial neural (MSDAN) network based on transfer learning is constructed to capture invariant features of human behaviors. To verify the superiority of PIDA, a series of experiments are conducted on various unseen domain data. The results showed that the PIDA method improves average DG accuracy by 1.52%, 2.48% and 4.88%, respectively, on the own dataset and 3.67%, 6.29% and 9.78%, respectively, on the public dataset, compared with baseline data, traditional optical data augmentation (TODA), and generative data augmentation (GDA) methods. Compared with current fall-related human behavior recognition works, our proposed method has a small DG gap.

In recent years, thermal imaging cameras are widely used in the field of intelligent surveillance because of their special imaging characteristics and better privacy protection properties. However, due to the low resolution and fixed location for current thermal imaging cameras, it is difficult to effectively identify human behavior using a single detection method based on skeletal keypoints. Therefore, a self-update learning method is proposed for fixed thermal imaging camera scenes, called the behavioral parameter field (BPF). This method can express the regularity of human behavior patterns concisely and directly. Firstly, the detection accuracy of small targets under low-resolution video is improved by optimizing the YOLOv4 network to obtain a human detection model under thermal imaging video. Secondly, the BPF model is designed to learn the human normal behavior features at each position. Finally, based on the learned BPF model, we propose to use metric modules, such as cosine similarity and intersection over union matching, to accomplish the classification of human abnormal behaviors. In the experimental stage, the living scene of the indoor elderly living alone is applied as our experimental case, and a variety of detection models are compared to the proposed method for verifying the effectiveness and practicability of the proposed behavioral parameter field in the self-collected thermal imaging dataset for the indoor elderly living alone.

Toward designing intelligent PDEs for computer vision: An optimal control approach

This paper presents data set of three commercial physiological sensors i.e. Electrocardiogram (ECG), Galvanic Skin Response (GSR), and Pulse sensor. The paper focuses on experiment to recognize human behavior using these body sensors. An experiment was done with participation of 12 users to observe human behavior i.e. Happy and Neutral mood. Users were asked to watch advertisement video based on comedy and actionscenes. During the implementation some variations were observed in the data-set while users were watching the videos. The results have been discussed at the end of the paper based on the data-set of 12 participants.

Soft Computing based object detection and tracking approaches: State-of-the-Art survey

Human detection and tracking is an interesting field of research in computer vision and image processing areas. It is widely used in video surveillance, robotics, human machine interaction and other applications. The automated object detection and tracking is still a challenging task that needs to be addressed. Hence the main idea is to develop a system based on various image processing techniques to reliably detect and track people in video sequence from stationary cameras. The proposed method can be viewed as consisting of two stages namely detection and tracking. In detection stage, Histogram of Oriented Gradients popularly known as HOG is used as a feature descriptor. HOG features are robust to local changes in geometry and illumination but it is computationally expensive. This disadvantage of HOG is due to its exhaustive scanning approach over entire region of interest. The main contribution of this paper is to reduce the computational time of HOG by dynamically determining the region of interest and limiting the scan area. Further Principal component analysis is used to reduce the dimensionality of HOG features. The additional use of optical flow based tracking eliminates the need of HOG computation in every frame. Hence the person detected by HOG in initial frame is successfully tracked in subsequent frames and reduces HOG computation time. The proposed dynamic ROI selection method not only reduces detection time but also reduces the number of false positives and increases the efficiency. The experimental results show that the system efficiently detects and tracks people in videos without much of occlusion.

Human behaviour recognition is one of the most fundamental tasks in Industrial Internet of Behaviour (IIoB) and is crucial for the safe and reliable IIoB. Existing methods lacks adaptability and transferability. In addition, there is a data isolation problem among different users. Therefore, there is an urgent requirement to construct a secure and adaptive human behaviour recognition model in IIoB without violating the privacy of users. Mamba, a structured state space model that integrates a selection mechanism and a scan module, is used for time series modelling tasks. To tackle the aforementioned problems, an Federated Learning-based lightweight human behaviour recognition model with selective state space models is proposed. First, we design a human behaviour recognition model integrating Mamba and residual structure to achieve lightweight and secure human behaviour modelling. In addition, considering data privacy and training efficiency, a decentralised dynamic FL framework is designed to achieve lightweight and secure model collaborative training, including: initial selection of source users, model aggregation strategy based on dynamic weighting, and fine-tuning module based on small-sample data, to improve the training efficiency of the model and the accuracy of human behaviour recognition. Extensive experiments are conducted to prove the superior performance of the proposed method.

With the continuous development of computer artificial intelligence technology, various applications based on artificial intelligence emerge in an endless stream, among which video image recognition technology is the most widely used in life. This article starts from the process of image recognition, based on the composite characteristics of artificial intelligence and video images, to discuss human gesture recognition technology.This article uses the feature extraction algorithm for image composite feature extraction as a method, and conducts human body movement collection experiments, analyzes the database and The gesture recognition step. This paper mainly introduces the extraction method of image composite features and the basic requirements of gesture recognition, and through the algorithm calculation of feature extraction, the function of human gesture recognition video and image composite features is completed, and the human action collection experiment is carried out to confirm. The results of images and data show the advantages of the algorithm support used in this article. We will Dmti. MsHOG is compared with other methods in the three subsets. In terms of the accuracy of all tests, our method performs better than other methods. The results show that the MSHOG (Multi-scale Histogram of Oriented Gradients) descriptor can represent the unique characteristics of human behavior, reflecting the effectiveness of our proposed method. In particular, this method achieved 100% recognition accuracy in Test, with an average recognition accuracy of 94.91%, which is significantly better than existing methods.

Neuroscience studies have shown that incorporating gaze view with third view perspective has a great influence to correctly infer human behaviors. Given the importance of both first and third person observations for the recognition of human behaviors, we propose a method that incorporates these observations in a technical system to enhance the recognition of human behaviors, thus improving beyond third person observations in a more robust human activity recognition system. First, we present the extension of our proposed semantic reasoning method by including gaze data and external observations as inputs to segment and infer human behaviors in complex real-world scenarios. Then, from the obtained results we demonstrate that the combination of gaze and external input sources greatly enhance the recognition of human behaviors. Our findings have been applied to a humanoid robot to online segment and recognize the observed human activities with better accuracy when using both input sources; for example, the activity recognition increases from 77% to 82% in our proposed pancake-making dataset. To provide completeness of our system, we have evaluated our approach with another dataset with a similar setup as the one proposed in this work, that is, the CMU-MMAC dataset. In this case, we improved the recognition of the activities for the egg scrambling scenario from 54% to 86% by combining the external views with the gaze information, thus showing the benefit of incorporating gaze information to infer human behaviors across different datasets.

Recently, device-free human behavior recognition has become a hot research topic and has achieved significant progress in the field of ubiquitous computing. Among various implementation, behavior recognition based on WiFi CSI (channel state information) has drawn wide attention due to its major advantages. This paper investigates more than 100 latest CSI based behavior recognition applications within the last 6 years and presents a comprehensive survey from every aspect of human behavior recognition. Firstly, this paper reviews general behavior recognition applications using the WiFi signal and presents the basic concept of CSI and the fundamental principle of CSI-based behavior recognition. This paper analyzes the key components and core characteristics of the system architecture of human behavior recognition using CSI. Afterward, we divide the sensing procedures into many steps and summarize the typical studies from these steps, including base signal selection, signal preprocessing, and identification approaches. Next, based on the recognition technique, we classify the applications into three groups, including pattern-based, model-based, and deep learning-based approach. In every group, we categorize the state-of-the-art applications into three groups, including coarse-grained specific behavior recognition, fine-grained specific behavior recognition, and activity inference. It elaborates the typical behavior recognition applications from five aspects, including experimental equipment, experimental scenario, behavior, classifier, and system performance. Then, this paper presents comprehensive discussions about representative applications from the implementation view and outlines the major consideration when developing a recognition system. Finally, this article concludes by analyzing the open issues of CSI-based behavior recognition applications and pointing out future research directions.

Computer vision and image processing technology have been rapidly developed and widely applied in many fields. There are many potential applications in modern agriculture. In this paper, a novel vegetable disease and insect pest recognition method is proposed based on the current computer vision and image processing methods. To investigate the vegetable disease and insect pest state, it is convenient to use images captured using smart phones for judgment. To implement this application, the disease area and the insect number on the leaves should be detected and figured out. So a new extraction and classification algorithm is firstly introduced to recognize leaves from images. Then a region-labeling algorithm is applied to calculate the insect number and disease areas in the segmented images. To deal with the areas of adhesion, a mathematical morphology algorithm is used for separating the objects. The proposed method is implemented on mobile smart devices and tested with field experiments. The experimental...