Coal And Coal Gangue Image Recognition Research Based On DSP

Coal is a kind of basic energy,accounts for about 70% in primary energy consumption structure,Coal resources in our country is very important for the economic development and people's life,In recent years,for the exploitation of the coal resources in our country still does not decrease,and quite a long period of time in the future,Coal resources will till holds the first place.There are gangues inevitably in mining.The gangue mixed in the coal,has low carbon content and low calorific value,not only reduces the quality of coal and reduces the combustion efficiency,but also releases harmful substances when burned that can aggravate environmental pollution.If you want to improve the utilization efficiency of coal,to realize separation of coal and gangue,therefore,the identification of coal and gangue is of important practical significance.Generally speaking,the methods of coal preparation are hand cleaning and machine cleaning.the main coal preparation methods are artificial coal,jigging coal,dry coal,?-ray coal and so on.But when these methods are used,the problems of resource waste,low sorting accuracy and high cost are coming.with the rapid development of vision technology,With these problems,people begin to use images to distinguish coal from gangue.Thus,a recognition method which is based on DSP image processing has been studied.First discusses identification scheme of coal and gangue,because it is based on digital image recognition method,need to find out the characteristics of the coal and coal gangue on the image,based on external color characteristics of coal and gangue,performance on the gray image is different grayscale distribution,so the two selected characteristic values of the mean gray as sorting according to the image characteristics of coal and gangue,developed a kind of image processing algorithms.First median filtering method was applied to suppress the image noise and smooth the edges of the image.Mapped the grayscale histogram of the image after processing,and then use Otsu method to image binarization,With morphological method for binary image processing,and then mapped to get coal and gangue target image on the original image;The final calculation of coal and gangue target image grayscale average.For identification of coal and waste rock is mainly use the difference of gray distribution of coal and gangue,so we first experiment by means of a series of samples of coal and gangue interval grey value distribution,and finally find the distinction between the two grayscale average threshold,will calculate the average compared with the threshold value to judge the picture like coal or coal gangue image.Finally in SEED-DEC6416 development board to realize image recognition algorithm,on the DSP operation speed,verify the feasibility of the proposed algorithm,the experimental results show that this method can well identify coal and coal gangue.