Search

Moving target detection and tracking, recognition, behaviors analysis are the key issues in the intelligent visual surveillance system (IVSS). The challenge is how to process the real-time video stream in an effective way in case that we could find the interested objects for analysis. However, the traditional video surveillance technology often does not meet the needs of real-time key frame recognition for the on-line intelligent video monitoring system. In our paper, we adopt the state-of-the-art Faster R-CNN [7] that takes advantages of convolutional neural networks into our real-time target recognition system - Deep Intelligent Visual Surveillance (DIVS). The key aspects of our DIVS are consisted of four parts: (i) Getting the real-time video images from remote cameras; (ii) Processing the data with the deep learning framework caffe [23] built for Faster R-CNN; (iii) Storing the valuable data with MySQL; (iv) Data presentation on the website. Experiments based on our system validated the effectiveness, stability and accuracy of our proposed solutions.

Route planning in current route guidance systems is primarily based on the shortest path strategy (e.g., suggesting a path that minimizes the travel distance between source and destination) for an individual driver. A drawback of such an approach is that there is a high probability that many drivers, who share the same source and destination, follow the same route. This can lead to over-saturated road links. We propose a simple yet effective multiple path routing algorithm that addresses this limitation. The proposed algorithm augments the A* search algorithm with a randomization component, which helps to perturb the order of the searched nodes and to produce diversified routes for different drivers with the same source and destination. Our algorithm enables decentralized route allocations, i.e., each vehicle computes its own route without a central server. We validate the performance of the proposed algorithm through microscopic traffic simulation. Experiments for various traffic scenarios show that our algorithm achieves a good path diversification while reducing the average travel time compared to the traditional shortest path algorithm.