红外技术, 2012, 34 (2): 99, 网络出版: 2012-04-20
移动机器人非结构场景图像边缘检测算法
Algorithm for Mobile Robot Unstructured Scene Image Edge Detection
摘要
针对非结构场景没有规则形状,传统算法不易识别以及红外图像对比度低很难区分灰度值相近的道路、水、天等目标的难点,提出了一种新的改进蚁群算法;新算法针对复杂背景图像的微弱边缘检测问题,围绕现有蚁群算法的关键步骤,从两个方面对其进行了改进,一是对信息素的表达式进行了改进,二是改进了启发信息值的计算方法;通过实例对新算法进行了验证,结果表明,新算法具有边缘检测位置准确、连续性好且干扰少的优点。
Abstract
Non-structural scene doesn’t have regular shape, and infrared image has low contract ratio. It’s hardly to identify objects with similar gray scale value, such as roads, water or sky, by traditional methods. According to this characteristic, considering weak edge detection under complicted background, this paper raised an improved ant colony algorithm, which improved ant colony algorithm in two aspects. The first improvement is expression of pheromone; the second one is calculation of heuristic information. This improved method is proved to be accurate in edge detection, good continuity and less interference by experiment.
王溪筠. 移动机器人非结构场景图像边缘检测算法[J]. 红外技术, 2012, 34(2): 99. WANG Xi-yun. Algorithm for Mobile Robot Unstructured Scene Image Edge Detection[J]. Infrared Technology, 2012, 34(2): 99.