A Revised Edge Detection Algorithm Based on Wavelet Transform for Coal Gangue Image

Based on wavelet transform theory, the digital image processing characteristics of coal gangues are studied to classify the coal gangues from coal bulk in the automation selecting system. Edge detection is a fundamental issue in the coal gangue image processing and pattern recognition. Through traditional methods of image processing are efficient as edge detectors, a lot of false edge information will be extracted from a coal gangue image, because the coal gangue images are always contaminated with speckles of coal dust. Therefore a revised algorithm based on wavelet transform is developed to suppress the speckles in the gangue images, to enhance edges. At last the gangues and coal are separated according to their characteristics. The experiment results show that the recognition rate of coal gangues is increased by the wavelet transform method.