摘要
提出了一种基于小波分解和分形纹理特征计算的图象分割方法。首先考虑对图象进行小波变换,然后对不同通道的子图象提取纹理的分形(分数维)特征和能量特征,最后用直方图阈值分割方法实现图象的分割。实验表明。
In this paper, an approach based on wavelet decomposition is developed for image segmentation. After making the image two\|dimensional wavelet transform, fractal dimension and texture features of subchannels are extracted for texture segmentation. The experimental results also demonstrated its satisfactory efficency both for simulated textured image and multispectral remotely sensed image.
出处
《中国图象图形学报(A辑)》
CSCD
1999年第12期1072-1077,共6页
Journal of Image and Graphics
基金
国家自然科学基金项目!(No.69675005)
关键词
图象分割
小波分解
分形
特征提取
遥感图象
Image segmentation, Texture, Wavelet decomposition, Fractal, Feature extraction