Gray Level Co-occurrence Matrix-based Edge Detection Algorithm

This article has carried on the analysis to the traditional gradient operator,and pointed out the existed drawback;At the same time,in view of the current texture analysis algorithm for ubiquitous complex compute and long operation time,through the combination of the texture analysis thought with the traditional gradient operator,the paper has proposed a new method of edge detection based on the gray level co-occurrence matrix.This method has maintained picture edge detail perfectly and noise restrained,because it has used the auto-adapted mask.The MATLAB simulation results show that the gray level co-occurrence matrix generated by the texture features can effectively describe the texture characteristics of fire with good differentiation.