The algorithm based on wavelet transform is presented to resolve detection of the water－sky－line polluted by heavy noise. Firstly, the image is wavelet transformed applying multiscale characteristics of wavelet. Then the output image with different scale is integrated into a simp le edge image by using multiscale edge match technique. As a result, the water－sky－line are extracted. The experimental results show that the algorithm can eliminate the strong clutter and noise, and is especially suited for the extraction of water－sky－line from the background of sea level.