Vision-Based Fall Detection Using BlazePose Skeletal Tracking and Bidirectional LSTM for Human-Centric Safety Systems

Health monitoring and tracking using Human Activity Recognition is a burgeoning study area, especially when it comes to detecting falls in the elderly. Falls can result in injuries, bodily injury, fractures, and other problems. Visual surveillance's major goal is to replace traditional video surveillance with intelligent visual surveillance. Different feature extraction algorithms were developed to better detect the fall. Convolutional Neural Network (CNN) in Fall Detection with the Optical Flow algorithm is used to extract features to find the fall detection. Designing CNN architecture for classification purposes without overfitting is a challenging task. The goal of this work is to avoid overfitting in training by automatically optimizing the CNN architecture with fine tuned dropout rate. This helps to increase the efficiency of the model. Four different CNN architecture are developed to predict fall detection. The UP-FALL Detection Dataset is used to test the performance of the four models. The experimental results show that second model produced high accuracy compared with other models.

Bicycles are affordable and healthy, cause zero harm to the environment, require little space on roads and parks, and are not noisy transportation means. However, cyclists face challenges on roads, which may lead to injuries and deaths. For instance, cyclists visibility, speed management, ability to communicate with other road users, backward accidents, and ability to be located in urgent cases. This paper presents performance, tracking, and safety systems from hardware and software point of view. As a result, cyclists monitor their speed to manage; their locations are tracked, and their cycling paths are traced. Also, a rear red light and turn signal LEDs are activated without any physical action from the rider to communicate with other drivers. Moreover, a microwave radar sensor with a range of seven meters is employed to protect the cyclist from backward-approaching vehicles. Finally, three Arduino UNO boards with GPS, GSM modules, MPU-6050, and RCWL-0516 sensors are mounted on a conventional bicycle to provide safer trips.

Falling can have fatal consequences for elderly people especially if the fallen person is unable to call for help due to loss of consciousness or any injury. Automatic fall detection systems can assist through prompt fall alarms and by minimizing the fear of falling when living independently at home. Existing vision-based fall detection systems lack generalization to unseen environments due to challenges such as variations in physical appearances, different camera viewpoints, occlusions, and background clutter. In this paper, we explore ways to overcome the above challenges and present Single Shot Human Fall Detector (SSHFD), a deep learning based framework for automatic fall detection from a single image. This is achieved through two key innovations. First, we present a human pose based fall representation which is invariant to appearance characteristics. Second, we present neural network models for 3d pose estimation and fall recognition which are resilient to missing joints due to occluded body parts. Experiments on public fall datasets show that our framework successfully transfers knowledge of 3d pose estimation and fall recognition learnt purely from synthetic data to unseen real-world data, showcasing its generalization capability for accurate fall detection in real-world scenarios.

Vision based fall detection solutions support elderly who live alone in their homes. For people falling down and not being able to stand up on their own again, such a fall is a major risk. In this work, we show a person detection approach using a relational feature model using consumer depth cameras. We propose a flexible relational feature model (FRFM) for fall detection in combination with histogram similarity functions such as the bhattacharyya distance, histogram intersection, histogram correlation and the chi-square 2 histogram similarity function. FRFM is an extension of the relational feature model (RFM) with the advantage that it can be used for all rotations of a body. The extension is necessary for fall detection due to the fact, that a lying person is rotated in the image where the standard person detection approach detects upright standing persons only. The relational features are computed for verification situations during fall detection on the basis of the Histograms of Oriented Gradients (HOG) feature descriptor. The experimental results show the best setup parameters for our feature model with different types of images (RGB and depth) and results on the specific field of fall detection.

The number of elderly people living alone has increased over the las1t years and fall is one of major risks that threaten their lives. A fall detection system has become a requirement and computer vision is an efficient solution among many accurate solutions developed in this field. This paper proposes a novel method vision-based fall detection using depth camera, which combines human shape analysis, head tracking and centroid detection to validate falls. An experimental test done with SDUFall dataset which contains 20 subjects performing five daily activities and falls demonstrates the efficiency of our method, achieving up to 93.25% accuracy compared with the state-of-the-art method using the same dataset.

Fall detection is one of the most important health care issues for elderly people at home, which can lead to severe injuries. With the advances and conveniences in computer vision in the last few decades, computer vision-based methods provide a promising way for detecting falls. In this paper, we propose a novel vision-based fall detection method for monitoring elderly people in house care environment. The foreground human silhouette is extracted via background modeling and tracked throughout the video sequence. The human body is represented with ellipse fitting, and the silhouette motion is modeled by an integrated normalized motion energy image computed over a short-term video sequence. Then, the shape deformation quantified from the fitted silhouettes is used as the features to distinguish different postures of the human. Finally, different postures are classified via a multi-class support vector machine and a context-free grammar-based method that provides longer range temporal constraints can verify the detected falls. Extensive experiments show that the proposed method has achieved a reliable result compared with other common methods.

Fall is an abnormal activity event in daily life, which is becoming a major cause of accidental death for the elderly. The purpose of fall detection is to minimize serious consequences and negative impacts after falling. Recently, most conventional vision-based fall detection methods mainly rely on hand-crafted features, which is inclined to be influenced by noises, etc. In this paper, a novel method is proposed to detect fall event during a video sequence by combining optical flow and wide residual network. The wide residual network contains fewer layers and achieves the same performance as residual network, which can make the neural network train faster. In addition, to model the video motion, we adopt optical flow images as input to the wide residual network. And finally, softmax classifier is utilized to distinguish fall event. The experimental results also confirm the fact that the proposed algorithm can achieve a reliable accuracy.

Workers might experience fall accidents even when they are working at flat level surfaces. These accidents plus other types of fall accidents have been reported as one of the major causes for worker-related fatalities and injuries. Currently, it becomes common to set up video cameras to monitor working environments. The video cameras provide an alternative to detect fall accidents. The objective of this paper is to investigate the feasibility of detecting fall accidents of workers with video. The preliminary focus is put on the fall detection under one single monocular camera. A novel fall detection method is proposed. Under the method, workers in the videos captured by the video cameras are first detected and tracked. Their pose and shape related features are then extracted. Given a set of features, an artificial neural network (ANN) classifier is further trained to automatically determine whether a fall happens. The method has been tested and the detection precision and recall were used to evaluate the method. The test results with high detection precision and recall indicated the method effectiveness. Also, the lessons and findings from this research are expected to build a solid foundation to create a vision-based fall detection solution for safety engineers.

Falls are one of the major causes of injury and death among elderly people aged 65 and industries too. A fall detection system is necessary to identify falling activities out of activities of daily lives. Even though new methods have shown on research paper, the number of studies in vision-based systems is still increasing. But existing vision-based fall detection systems have lot of weakness to be generalized mainly due to the difficulties such as variations in physical appearances, different camera viewpoints, occlusions, background clutter and darkness. Head and upper body location of human provides a critical information at the time of fall. This paper presents a vision-based fall tracking method where upper body joints are grouped into one segment to increase the fall classification ratio. Segment consist of Head and shoulders joints combinations represents the upper region and head region. Segment method can be beneficial to achieve an efficient tracking of human activities and provide strong technic to distinguish falls from activities of daily lives.

Falls are the leading cause of accidents among the elderly population. In recent years, radar has been employed in fall detection due to its superior sensing capabilities, small dimensions, low cost and primarily non-intrusive sensing capabilities in addition to its robustness under a range of heat and lighting conditions. In this paper, we present a technique for identifying when a person is falling using a low-power millimeter-wave radar operating in the W-band. This detection, conducted in real time, is based on the transmission of a continuous wave and heterodyning of the received signal reflected from the person to obtain micro-Doppler shifts associated with the person’s motion. These results make it possible to obtain a high-quality time-frequency distribution and spectrogram, from which the person’s unique fall movement characteristics can be determined. In this paper, we present experimental results based on 94 GHz real radar data obtained from a falling person. This carrier frequency is higher than that of current systems, allowing higher frequency resolution and more accurate results. Compared to other tracking systems, this sensor does not simulate or violate privacy. However, the high-frequency system enables high-resolution realizations with high reliability.

<p>The charity box is a supporting facility that can be used by pilgrims to distribute their infaq. In this modern era, it turns out that the mosque still counts the contents of the charity box manually and uses the usual padlock key for the safety of the charity box. The purpose of this research is to build a system and tool that can simplify the performance of mosque administrators in counting money and maintaining the security of the charity box. ArduinoBased Charity Box Safety, Tracking, and Counter System is a charity box equipped with automatic counting and security features using RFID and GPS. The method used in making this system is the waterfall method. The features in this system are automatic counting and security features using RFID keys, GPS, buzzers, and infrared sensors. This system is also equipped with a notification to the mosque management regarding the amount of money in the charity box and the location of the charity box using GSM / GPRS. From the testing of the system, it was produced that the Arduino-Based Charity Box Safety, Tracing, and Counter System was able to detect banknotes and coins well. The SMS feature also works well where there will be an SMS message regarding the amount of money and location of the charity box. The safety of the charity box is enhanced by using RFID, GPS, buzzers, and infrared sensors. From the user satisfaction test results obtained by the percentage of user satisfaction by 85%, which means the Arduino-Based Charity Box Safety, Tracing, and Counter System is quite attractive to users.</p>

Recognition of human movements is very useful for several applications, such as smart rooms, interactive virtual reality systems, human detection and environment modeling. The objective of this work focuses on the detection and classification of falls based on variations in human silhouette shape, a key challenge in computer vision. Falls are a major health concern, specifically for the elderly. In this study, the detection is achieved with a multivariate exponentially weighted moving average (MEWMA) monitoring scheme, which is effective in detecting falls because it is sensitive to small changes. Unfortunately, an MEWMA statistic fails to differentiate real falls from some fall-like gestures. To remedy this limitation, a classification stage based on a support vector machine (SVM) is applied on detected sequences. To validate this methodology, two fall detection datasets have been tested: the University of Rzeszow fall detection dataset (URFD) and the fall detection dataset (FDD). The results of the MEWMA-based SVM are compared with three other classifiers: neural network (NN), naive Bayes and K-nearest neighbor (KNN). These results show the capability of the developed strategy to distinguish fall events, suggesting that it can raise an early alert in the fall incidents.

A Vision Based approach for Fall Detection System for Elderly People

The use of millimeter-wave (mmWave) frequencies is a promising technology for meeting the ever-growing data traffic in next-generation wireless communications. A major challenge of mmWave communications is the high path loss. To overcome this issue, mmWave systems adopt beamforming techniques, which require robust channel estimation and tracking algorithms to maintain an adequate quality of service. In this study, we propose the machine learning-based channel tracking algorithm for vehicular mmWave communications. In this paper, we propose a long short-term memory (LSTM)-based channel tracking algorithm for vehicle-to-infrastructure mmWave communications. The bidirectional LSTM is leveraged to track the channel. Simulation results demonstrate that the proposed algorithm efficiently tracks the mmWave channel with negligible training overhead.

Accidental falls of elderly people are a major cause of fatal injuries, especially for those living alone. We present a novel vision–based fall detection approach that analyzes an extracted human body using described human postures. First, a human body extracted by a background subtraction technique is located by a minimum area-enclosing ellipse. Then, a normalized directional histogram is developed around the center of the ellipse to represent a human posture by multi-directional statistical analysis. After that, 12 static and 8 dynamic features are derived from the normalized directional histogram. These features are fed into a directed acyclic graph support vector machine to distinguish four closely related human postures (standing, crouching, lying, and sitting). A fall-like accident is detected by counting the occurrences of lying postures in a short temporal window. After conducting majority voting, a fall event is determined by immobility verification. From the experimental results, an overall accu...