基于麻雀优化的二维Otsu煤岩界面识别方法

Two-dimensional Otsu coal-rock interface recognition method optimized by sparrow search algorithm

由于传统的煤岩界面识别方法识别精度较低,导致煤炭资源的采出率降低,煤中含矸量提高,影响工业经济的发展。针对上述问题,文中设计了一种基于麻雀搜索算法优化的二维Otsu煤岩界面图像识别方法,来实现煤岩界面的精准识别。以陕西神木榆家梁采煤工作面的图像为例进行研究。首先将工作面原图像进行图像去噪和图像增强预处理;接着利用文中所设计的算法对预处理后图像进行分割,并将该方法分别与传统二维Otsu算法和粒子群优化的二维Otsu算法进行比较。结果表明:文中设计算法较传统方法在图像分割的运行速度和精度方面都有了较大提升;利用Canny算子检测分割后图像的边界并通过形态学操作处理,得到了与真实的煤岩界面基本吻合的煤岩分界线,验证了该方法的可行性和有效性。

Due to the low recognition accuracy of the traditional coal-rock interface recognition methods,the mining rate of coal resources is reduced while the amount of gangue in coal is increased,which affect the development of industrial economy.Therefore,a two-dimensional Otsu coal-rock interface image recognition method optimized by sparrow search algorithm(SSA)is designed to realize the accurate recognition of the coal-rock interface. The image of the working face of coal mining in Yujialiang Coal Mine, Shenmu, Shaanxi Province is taken as an example. The original image of the working face is subjected to preprocessing of image denoising and image enhancement. The designed algorithm is used to segment the preprocessed image,and then is contrasted with the traditional two-dimensional Otsu algorithm and the two-dimensional Otsu algorithm optimized by the particle swarm optimization(PSO) algorithm respectively. The results show that the designed algorithm has a greater improvement in the running speed and accuracy of image segmentation in comparison with the traditional methods. The Canny operator is used to detect the boundary of the segmented image, and then the boundary is processed by morphological operations,so as to obtain the coal-rock boundary line basically consistent with the real coal-rock interface. Therefore,the feasibility and effectiveness of the method has been verified.