基于DSP与改进边缘检测算法的煤矸石自动分选系统

摘要

The system investigated the fundamental characteristics of coal images and gangue images on the basis of digital image processing technology. Median filter, edge extraction and gray comparison of coal images and gangue images were successively carried out by the technology of image preprocessing, image segmentation, feature extraction and image recognition. The image processing results showed that coal images and gangue images could be distinguished based on their gray histogram. Coal gangue could been separated from the lump coal. To improve operation speed, the high performance microcontroller was used as the core processor of system. And the improved edge detection algorithm was also used in the system. It can achieve the real-time automatic separation of coal gangue.%利用数字图像处理技术分析煤块图像和矸石图像的基本特征.通过图像预处理、图像分割、特征提取和图像识别等图像处理技术,依次对煤块图像和矸石图像进行中值滤波、边缘提取和灰度比较等分析.图像处理结果表明:根据各自的灰度直方图特性,区分出煤块图像和矸石图像,进而分选出煤块和矸石.采用高性能的数字信号处理器(DSP)作为系统的核心处理器,提高了运算速度,并结合改进的边缘检测算法,实现了煤块和矸石的实时自动分选.