To eliminate the problems of extracting false corners and losing the information of real corners and to overcome the difficulty in finding a universal threshold in the non-maximal inhibition for the processing of all pictures in the Harris corner detection algorithm, an autoadaptive threshold is adopted to generate more accurate corners. To further reduce false corners and accelerate the processing of large images, a pre-selection strategy is embedded to rule out the pixels which are definitely not corners and pick out potential corners before the normal routine. This implementation reduces the amount of calculation, thus enhances the efficiency of the Harris algorithm. Experimental results show that the improved algorithm can detect more accurate corners and fewer false corners in shorter than 30％ of the processing time required by the original Harris algorithm.%为克服Harris角点检测算法中漏检正确角点和提取出较多伪角点的问题,以及在对不同图像处理时,非极大值抑制无法设置通用阈值的现象,本文在进行非极大值抑制时采用自适应阈值,从而可得较多的正确角点.为进一步抑制角点检测中的伪角点数目并提高处理大图像的算法效率,加入预筛选得备选角点这一步骤,通过在进行Harris角点检测之前就先去除部分肯定不可能是角点的像素点,以减少最终得到的伪角点数,并有效地减少了运算量,提高了效率.实验结果显示改进的Harris角点检测算法的运行时间仅为原始算法的30%,且可以得到更多的精确角点和更少的伪角点,具有很好的角点检测性能.
展开▼
