【摘要】 研究了一种基于harris(哈里斯)角点匹配的电子稳像算法。为了实现电子稳像,该算法首先从震动图像中搜索所有的哈里斯角点,然后选择左上角的角点作为模板,在震动前的图像中寻找匹配部分,从而计算出运动向量,最后根据该运动向量计算出稳定的图像。结果表明,该算法是一种快速、高效而且稳定的方法,在井下智能监控系统的应用中能取得良好的效果。更多还原

【Abstract】 Electronic image stabilization(EIS) is an important step of many systems of computer vision.This paper researches the algorithm of EIS based on harris corners-matching method.In order to achieve EIS,the algorithm first searches all harris corners from the motion picture,then selects the upper-left corner as the template and finds the matching part from the early picture to calculate motion vector,and at last the stabilized image is calculated according to the motion vector.The result demonstrates that the algorithm is a fast,efficient and stable method.It has achieved good results in the application of the Underground Intelligent Monitoring System.更多还原