Accident Detection System for Bicycle Riders

Abstract

Bicycle riders are exposed to accident injuries such as head trauma. The risk of these riders' injuries is higher compared to the risk of injuries for motorists. Crashes, riders' errors, and environmental hazards are the cause of bicycle-related accidents. In 2017, nearly 50% of bicycle-related accidents occurred in urban areas at night, which may contribute to a delay in reporting the accidents to emergency centers. Hence, a system that can detect the accident is needed to notify urgent care clinics promptly. In this article, we propose a bicycle accident detection system. We designed hardware modules measuring the features related to the riding status of a bicycle and fall accidents. For this purpose, we used a magnetic, angular rate, and gravity (MARG) sensor-based system which measures four different types of signals: 1) acceleration, 2) angular velocity, 3) angle, and 4) magnitude of the riding status. Each of these signals is measured in three different directions (X, Y, and Z). We used two different time-domain parameters, i.e., average and standard deviation. As a result, we considered 24 features. We used principal component analysis (PCA) for feature reduction and the support vector machines (SVM) algorithm for the detection of fall accidents. Experimental results show that our proposed system detects fall accidents during cycling status with 95.2% accuracy, which demonstrates the feasibility of our proposed bicycle accident detection system. Index Terms-Bicycle accident, fall detection, MARG, principal component analysis, support vector machines. I. INTRODUCTION A CCORDING to the Centers for Disease Control and Prevention's (CDC) report published in 2017 [1], bicycle-related accidents lead to around 1,000 fatal and 467,000 non-fatal injuries. There are various causes associated with these bicycle-related accidents such as rider's carelessness, environmental hazards, and crashes with motor vehicles or other bicycles [2]. More than 45% of bicycle-related accidents are reported to occur in dark conditions and urban areas [1], [3]. Moreover, in some situations, the rider may be alone or unable to ask for help due to shock, injuries, or unconsciousness, which can delay medical treatment [4]. There have been studies on analyzing and monitoring riding conditions to increase the safety of bicycle riders [3], [5]-[10]. In [9], a smartphone is used to recognize four different types of riding status-right turn, left turn, straight run, and stop. Here, the smartphone was attached to a bicycle handlebar and collected accelerometer and gyroscope data to recognize the movement type of the bicycle based on Manuscript

A Study on the Application of LSTM to Judge Bike Accidents for Inflating Wearable Airbags

“…Figure 12 shows the receiver operating characteristic (ROC) curves of LSTM, CNN, and NN. The ROC curve is a curve representing the performance of how well the class of binary classification can be distinguished [ 22 ]. For example, when an accident situation is set to 1, and a non-accident situation is set to 0, there are four cases: 1 is judged as 1 (true positive), 0 is judged as 1 (false positive), 1 is judged as 0 (false negative), and 0 is judged as 0 (true negative).…”

“…The NN had the same structure as the NN used in the LSTM model. The CNN used the structure used in the precedent studies [ 20 , 21 , 22 , 23 ]. The convolution layer used 8 sets of 15 × 1 × 1 three-dimensional matrices as a kernel, and the stride was composed of a 1 × 1 × 1 matrix.…”

“…A study by [ 21 ] provided an analysis of the accident sequence in the event of a two-wheeled vehicle accident. Another study [ 22 ] presented a magnetic, angular rate, and gravity sensor-based system that detects accidents with 95.2% accuracy through principal component analysis (PCA) and support vector machines (SVMs) and proposed a system that can be notified to emergency medical centers. In [ 23 ], the authors presented an idea for an efficient design method of a detection system that effectively uses a large number of sensors through a two-step detection algorithm based on the self-organizing maps (SOMs) method.…”

Wound Healing: From Passive to Smart Dressings

“…Therefore, various innovative wound dressings with antibacterial capacity, including paper‐based dressings, microneedle patches, and flexible microelectronic integrated skin bandages for real‐time monitoring and treating wound bacterial infection, have been developed. [ 106,108 ]…”

Development and Evaluation of a Virtual Environment to Assess Cycling Hazard Perception Skills

“…Hazard perception can be defined as the ability to anticipate dangerous situations based on perceptual evidence [ 4 ]. Research suggests that expertise differences in hazard perception range from solely identifying salient information (e.g., cars) to “reading” more complex travel scenes to evaluate potential dangerous scenes of a less salient nature (e.g., considering not having a line of sight and covert hazards, [ 5 , 6 ]). These hazard perception abilities are typically examined in video-based paradigms, where reaction times, response rates, and eye movements from a road user’s perspective are investigated.…”