Fall Detection Based on Dual-Channel Feature Integration

Abstract

Falls have caught great harm to the elderly living alone at home. This paper presents a novel visual-based fall detection approach by Dual-Channel Feature Integration. The proposed approach divides the fall event into two parts: falling-state and fallen-state, which describes the fall events from dynamic and static perspectives. Firstly, the object detection model (Yolo) and the human posture detection model (OpenPose) are used for preprocessing to obtain key points and the position information of a human body. Then, a dual-channel sliding window model is designed to extract the dynamic features of the human body (centroid speed, upper limb velocity) and static features (human external ellipse). After that, MLP (Multilayer Perceptron) and Random Forest are applied to classify the dynamic and static feature data separately. Finally, the classification results are combined for fall detection. Experimental results show that the proposed approach achieves an accuracy of 97.33% and 96.91% when tested with UR Fall Detection Dataset and Le2i Fall Detection Dataset. INDEX TERMS Computer vision, fall detection, dual channel, machine learning, classification.

Comprehensive Review of Vision-Based Fall Detection Systems

“…In [ 87 ], the object detection task, performed by a YoLO CNN, is combined with object tracking, a task developed by DeepSORT [ 129 ], a CNN architecture able to track multiple objects after they have been detected.…”

“…Certain convolutional architectures, like the ones integrated into OpenPose and used in [ 87 , 90 ], can identify human body key points through convolutional pose machines (CPM), as shown in Figure 3 , a CNN able to identify those features. These key points are used to build a vector model of the human body in a bottom-up approach.…”

“…To reduce the complexity of the trees, and therefore, their computational cost, branches are pruned when the relation cost-saving/accuracy loss is satisfactory. This type of classifier is used in [ 87 , 89 ].…”

“…Random forest (RF), like the one used in [ 54 , 87 ], is an aggregation technique of DT, introduced by Braiman [ 139 ], which main objective is avoiding overfitting. To accomplish this task, the training dataset is divided into subgroups, and therefore, a final number of DTs, equal to the number of dataset subgroups, is obtained.…”

“…Multilayer perceptron (MLP) is a type of multilayered ANN with hidden layers between the entrance and the exit ones able to sort out classes non linearly separable. Each node of this network is a neuron that uses a nonlinear activation function, and it is used for classification purposes in [ 48 , 87 ].…”