Image edge detection is a core step in image processing.The edge detection results directly affect the im?age detection results, but there is lacking a standard numerical evaluation measure of the edge detection results. Hence, this paper proposes an algorithm for reconstructing the original image with an edge image. The algorithm searches the pixels of the original image in latitudinal and longitudinal directions and then reconstructs new pixels by linear interpolation and gradient interpolation between new and original pixels.Finally it evaluates the edge detec?tion effects with the reconstructed similarity between new and original images. The experimental results show that the algorithm can quickly and effectively reconstruct the original image with the edge image and can effectively eval?uate the edge detection effects. The image reconstruction results and evaluation results conform to human visual per?ception. Therefore the algorithm is applicable to high?level and automatic image processing.%边缘检测是图像处理的一个核心步骤,其检测效果直接决定图像处理的质量,但对于图像边缘检测一直缺乏量化的标准评价方式.文章提出了一种用边缘图像重构源图像的算法,横纵多方向搜索重构源集,以线性插值与渐变插值混合重构新像素,用重构图像与源图像的相似性评价边缘检测效果.设计了多种实验,对重构算法的性能与效果进行验证.实验结果表明,文中所构造算法能快速、有效地以边缘图像重构源图像,且可有效评价边缘检测效果.重构结果与评价结果符合人的视觉感知,对于高层次的图像处理与自动化图像处理具有较好的应用价值.
展开▼
