在图像测量等工程应用中,需要获得目标的高精度图像边缘信息.本文首先介绍了一种基于Facet模型的亚像素边缘检测算法,该方法具有抗噪能力强,定位精度高等优点,但是计算复杂度太高.针对此缺点,本文研究了一种改进算法,将其与Mallat的小波变换模极大算法有效的结合,不仅处理速度提高了10倍左右,而且所提取的边缘效果也有所改善.实验结果表明,本文的方法不仅能获得高精度的边缘信息,抗噪能力强,而且处理速度快。-Need to obtain the target image measurement and other engineering applications, high-precision image edge information. This paper first introduces a sub-pixel edge detection algorithm based on the Facet model, the noise immunity, high precision positioning, but the computational complexity too high. this shortcoming, this paper, an improved algorithm, Mallat wavelet transform modulus maxima algorithm effective combination, not only the processing speed of about 10 times, and extracted the edge effect is also improved. The experimental results show that this method can not only get to the edge of the high-precision, noise immunity, and processing speed.