Objective Recognition Algorithm of Remote Sensing Image Based on Wavelet Transform
Ben Tu Li, Zhi Chao Yuan · Applied Mechanics and Materials · 2013
The traditional template matching algorithm which has a high time complexity, is susceptible to noise. This paper proposed a algorithm which based on the features of continuous orthogonal wavelet Daubechies, decomposed images to be identified to multiple layers using wavelet tranformation. In order to get matching position, we select matching template in low-frequency images. Then get matched position in higher layers after doing inverse transformation to low-frequency image. Finally, accurate position of matching template will get in original images. The algorithm not only can reduce the searching time when images are matched, but also can filter out a certain amount of noise, and so reduce the noise interference.