论文部分内容阅读
本文首先研究了Gabor滤波器进行确定性纹理分割时纹理结构与Gabor函数的关系，提出了利用在频域中具有紧支集的Shannon小波包分解检测纹理的主频;最后.提出了基于小波包分解自适应Gabor函数设计的纹理分割算法.该算法首先用Shannon小波包分解检测纹理的主频，求设计Gabor函数，然后根据Gabor函数与纹理图象卷积、就可以在纹理的连接处产生良好的阶跃边缘，并通过实验验证了该算法的有效性及其对噪声的鲁棒性。
In this paper, the relationship between the texture structure and the Gabor function in the Gabor filter for deterministic texture segmentation is studied first, and the dominant frequency of the texture is detected using Shannon wavelet packet decomposition with compact support in the frequency domain. Finally, The algorithm decomposes the adaptive Gabor function to design the texture segmentation algorithm. The algorithm firstly uses the Shannon wavelet packet decomposition to detect the dominant frequency of the texture, and then designs the Gabor function. Then, according to the Gabor function and texture image convolution, The experimental results show that the algorithm is effective and robust to noise.