【摘要】 分析并设计了具有不随目标的平移、旋转和尺度变换以及边界的起点变化而变化的傅里叶描述子。把小波变换的多尺度边缘检测算法及傅里叶描述子应用于基于内容的图像检索。实验结果表明 ,这种方法对于有背景噪声和旋转、平移及尺度变换后的同一类图像的检索率是很高的更多还原

【Abstract】 In this paper, a multi-resolution edge detection arithmetic based on wavelet transform is applied to image with noise. It has a better performance than classical edge detection operators. Fourier descriptors which do not change with translation, rotation and scale change are analyzed. Then multi-resolution edge detection arithmetic based on wavelet transform and Fourier descriptors are applied to shape-based image retrieval. The experimental result proves that retrieval ratio is high in case of background rotation, translation and scale change.更多还原