一种基于小波变换的煤矸石图像边缘检测方法

Edge detection based on wavelet transform for coal gangue image

针对煤矸石在线识别与自动分选系统中的图像处理问题,采用小波变换的多尺度方法,研究了煤矸石图像的轮廓提取与边缘检测,通过图像消噪和边缘加强,对煤矸石图像进行了处理。研究结果显示,采用小波变换的边缘检测方法,可以有效地抑制噪声,准确提取煤矸石图像的边缘特征,有助于提高煤矸石的识别率。

In allusion to the existing problems on coal gangue image processing in the on-line recognition and automation selecting system, an algorithm of edge detection based on wavelet transform multi-resolution analysis is proposed to research on the profile extraction and edge detection. Through reducing of the speckles of coal dust and edge enhancement, the coal gangue images are processed according to their characteristics. Investigation results show that the method not only effectively reduces noises, but also accurately extracts the image edge, and the recognition rate of coal gangues is increased by the wavelet transform.