PROJECT TITLE :

Wearable Camera- and Accelerometer-Based Fall Detection on Portable Devices

ABSTRACT:

Robust and reliable detection of falls is crucial particularly for elderly activity monitoring systems. During this letter, we gift a fall detection system using wearable devices, e.g., smartphones, and tablets, equipped with cameras and accelerometers. Since the moveable device is worn by the topic, monitoring is not limited to confined areas, and extends to wherever the topic might travel, vs static sensors installed in bound rooms. Moreover, a camera provides an abundance of information, and also the results presented here show that fusing camera and accelerometer information not solely increases the detection rate, however also decreases the number of false alarms compared to solely accelerometer-primarily based or only camera-based systems. We tend to employ histograms of edge orientations together with the gradient local binary patterns for the camera-based mostly half of fall detection. We compared the performance of the proposed technique with that of using original histograms of oriented gradients (HOG) also a modified version of HOG. Experimental results show that the proposed technique outperforms using original HOG and changed HOG, and provides lower false positive rates for the camera-based mostly detection. Moreover, we have a tendency to have utilized an accelerometer-based mostly fall detection technique, and fused these two sensor modalities to own a strong fall detection system. Experimental results and trials with actual Samsung Galaxy phones show that the proposed method, combining two completely different sensor modalities, provides abundant higher sensitivity, and a significant decrease in the amount of false positives throughout daily activities, compared to accelerometer-solely and camera-solely ways.