The Simulation Application of Wavelet Transform for Harbor Detection in Map

The corner points are the most significant characteristic of the harbor in coast line, and corner points detection has some classic algorithm, such as using Gaussian smoothing filter, but there is phenomenon of corner information missing and corner migration with this method. In order to detect the harbor of coast line in the map exactly, in this paper, a new algorithm based on the wavelet transform is proposed to detect the corner points of harbor. Firstly, the map is preprocessed by the morphologic algorithm, then the sobel operator is used to detect the coast line of the map. Weed out the shorter curves then the longest curve is retained as the curve of the coast line, and transform it into the angle function, then calculate the convolution with the wavelet function in four different scales(k = 0, 1/2, 1, 3/2). The extreme points of the convolution results are taken as the corner points of the harbor in map, then the harbor is detected. The simulation results show the validity of the proposed algorithm.