论文部分内容阅读
输电线路视频图像中的间隔棒是获得导线状态信息的重要元素。为了实现对输电线路图像中间隔棒的快速、准确定位,提出了一种基于梯度方向场优化的快速模板匹配算法。该算法针对传统的相关匹配算法运算量大、直方图匹配结果不准确的缺陷,提出了直方图和归一化相关匹配结合的复合匹配方法。结合输电线路图像中导线方向明确的特点,应用梯度方向场对图像中输电线路目标区域进行锁定,减少模板的搜索空间。实验结果表明,该算法具有较强的鲁棒性,在保证匹配精度的同时,大幅提高了匹配速度,能够满足远程视频监控系统对电力设备图像识别实时性和准确性的需求,可为准确判定输电线路状态信息提供依据。
Spacers in a video image of a transmission line are important elements in obtaining information on the state of a conductor. In order to achieve fast and accurate positioning of spacers in transmission line images, a fast template matching algorithm based on gradient direction field optimization is proposed. Aiming at the defect that the traditional correlation matching algorithm has a large amount of operation and inaccurate histogram matching, this algorithm proposed a compound matching method combining histogram and normalized correlation matching. Combined with the characteristics of the clear line in the transmission line image, the gradient direction field is used to lock the target area of the transmission line in the image and reduce the searching space of the template. Experimental results show that this algorithm has strong robustness, greatly improves the matching speed while ensuring matching accuracy, and can meet the real-time and accuracy requirements of remote video surveillance system for power equipment image recognition, which can be used to accurately determine Transmission line status information provided basis.