Accurate Vehicle Detection Using Multi-Camera Data Fusion and Machine Learning

Computer-vision methods have recently been extensively used in intelligent transportation systems for vehicle detection. However, the detection of severely occluded or partially observed vehicles due to the limited camera fields of view remains a significant challenge. This paper presents a multi-camera vehicle detection system that significantly improves the detection performance under occlusion conditions. The key elements of the proposed method include a novel multi-view region proposal network that localizes the candidate vehicles on the ground plane. We also infer the vehicle occupancies by leveraging multi-view cross-camera context. Experiments are conducted on a dataset captured from a roadway in Richardson, TX, USA, and the proposed system attains 0.7849 Average Precision (AP) and 0.7089 Multi Object Detection Precision (MODP). The proposed system advances the single-view region proposal approaches by approximately 31.2% for AP and 8.6% for MODP