期刊刊名：黎明學報 　 卷期：28卷1期

篇名出版日期：2017年1月31日

作者：Li-Chih Chen,陳勵志

語言：English

關鍵字：surveillance systems, vehicle make and model recognition (MMR), symmetrical SURF,車輛監控系統,車型辨識,對稱SURF

被點閱次數：1次

閱讀時間：1sec

摘要：
In computer vision, the vehicle detection and identification is a very important research topic. In the intelligent vehicle monitoring analysis application system, in order to obtain the correct vehicle-related information, it must first be able to detect ROI (Region of Interest) of vehicle exactly. To detect and analyze the information of vehicles is one of important issues in security surveillance and intelligence transportation system application. Nowadays, there are many cameras and surveillance systems mounted in the intersections of cities to monitor passing vehicles. This paper adopts symmetrical SURF descriptor which enhances the ability of SURF to detect all possible symmetrical matching pairs for vehicle detection and analysis. Each vehicle can be found accurately and efficiently by the matching results even though only single image without using any motion features. Our proposed detection scheme has two main advantages, i.e., (1) no need using background subtraction method and (2) efficiency for real-time applications. After that, one challenging task is vehicle make and model recognition (MMR), there are two challenging tasks in MMR, i.e., the multiplicity and ambiguity problems. The multiplicity results from the slight modification of same type vehicle by the manufacturer and the ambiguity stems from different type vehicles which often share similar shapes. To tackle these two problems, we adopt a grid division scheme to divide a vehicle into several grids. Then, these different weak classifiers are trained individually, and then combining these weak classifiers builds a strong ensemble classifier. The ensemble classifier can accurately recognize each type vehicle. Experimental results prove the superiorities of our method in vehicle MMR.

　　　　　[全文下載]