摘要
提出了利用灰度图像边缘检测结果完成彩色图像边缘检测的算法,并对Canny算子进行了改进,在计算梯度幅值时采用了3×3邻域,改进后的Canny算子对噪声抑制效果明显,且检测边缘更加精细.采用了两种算法将彩色图像转换为灰度图像,利用C-Canny算子对转换后的灰度图像进行边缘检测,在灰度图像边缘检测结果基础上,完成彩色图像的边缘检测.经实际测试证明,该算法检测出的图像边缘连续性好,准确性高,对颜色数不多的彩色图像有较好的边缘检测结果.
Based on gray space and C-Canny operator, an edge detection algorithm of color image is proposed. And the Canny operator has been improved, using 3 × 3 neighborhood to calculate the gradient magnitude. The improved Canny operator gets a good effect on the noise suppression, and the detected edges are much more refined. Two transformation algorithms of color image to gray image are used. Execute C-Canny operator on the transformed image, and the result will help us complete edge detection on the color image. Based on our algorithm, the experiment showed that we will get good continuity and high veracity of image edge, and better edge detection will obtained on the color image of small number of colors.
出处
《微电子学与计算机》
CSCD
北大核心
2010年第4期17-20,共4页
Microelectronics & Computer
