Abstract: Edge detection is the base of the image measuring，image segmentation，image compressing and pattern recognition and so on.Color images provide more information than monochrome image，and color image edge detection has been paid more and more attentions.After briefly reviewing the existing color image edge detection methods，color image edge detection based on principal axis analysis and embedded confidence is proposed.The experimental results show that this algorithm applies the color feature fully，is rich in feature content and accurate and obtains a better edge continuity.

Key words: principal axis analysis, embedded confidence edge detection, color image edge detection

摘要： 边缘提取是获取图像特征的基本方法之一，彩色图像提供了比灰度图像更丰富的信息，彩色图像的边缘检测日益受到人们的重视。分析了目前常见的彩色图像边缘检测算法，提出了一种将主轴分析和嵌入置信度相结合的边缘提取算法，通过实验证明，该方法充分利用了图像的彩色信息，能够有效地保护边缘细节，提高检测精度，具有良好的边缘提取效果和边缘连续性。

关键词: 主轴分析法, 嵌入置信度的边缘检测, 彩色图像边缘检测