针对墙体裂缝对比度低、受噪声干扰严重等特点,提出了一种基于张量投票和目标追踪的墙体裂缝检测算法。在对原始影像进行二值化的基础上,利用张量投票的聚类思想,突出线性区域目标的强度,然后在突出的线性目标上选取特征点,对特征点进行追踪得到裂缝的大致位置,再反投到原始影像上。实验结果表明,该算法能够正确检测出墙体裂缝信息,且检测率较高,具有一定的通用性。 In view of the low contrast of wall cracks and the serious noise interference, a crack detection algorithm based on tensor voting and target tracking is proposed. Based on the binarization of the original image, the tensor voting clustering is used to highlight the intensity of the target in the linear region. Then the feature points are selected on the prominent linear targets, the feature points are traced to obtain the approximate position of the crack, Re-cast to the original image. Experimental results show that this algorithm can correctly detect the wall crack information, and the detection rate is high, with a certain versatility.