【摘要】 本文分析了目前常见的亚像素边缘检测方法,提出了一种综合运用主轴分析法、Zernike矩和嵌入置信度的彩色图像亚像素边缘检测算法。首先获取影像三维空间中的主轴,然后利用嵌入置信度的边缘检测算子快速检测出所有可能的边缘点,再利用Zernike矩算子以亚像素精度重新定位边缘。实验表明本文算法优于其它的边缘检测算子,充分利用了图像的彩色信息,定位精度能够达到0.13个像素,并且该方法还具有良好的抗噪声干扰性和处理效率。更多还原

【Abstract】 This paper reviews the existing sub-pixel edge detection methods and presents a new edge detection meth- od with sub-pixel accuracy for color images,which uses the principal axis analysis,Zernike moment and embedded confidence.First,the principal axis of the color vectors in three-dimensional space is obtained from the image through analysis.Then all the possible edge points are quickly detected by the edge detection operator with embed- ded confidence.At last the edge is relocated with sub-pixel accuracy by means of Zernike moment operator.Experi- ments show that the proposed method fully utilizes the color feature of the image;the accuracy of the method is supe- rior to other edge detectors and the location accuracy reaches to 0.13 pixels.Besides,this method also has strong anti-interference capability and high processing efficiency.更多还原