带式输送机托辊红外图像分割与定位算法

发布时间：2018-05-16 01:32

本文选题：带式输送机 + 红外图像　；参考：《西安科技大学学报》2017年06期

【摘要】：由于煤矿带式输送机关键部件缺乏有效监测,而传统目检、温度监测方法存在工作量大、盲点多等问题,文中提出一种基于连通分量的带式输送机托辊红外图像自动分割与定位算法,对巡检机器人沿巡检轨道采集的带式输送机红外图像进行处理,利用垂直和水平投影截取托辊所在区域,减少支架、输送带以及背景对后续图像处理的影响;采用基于连通分量的长短轴比和面积信息对图像边缘进行过滤,消除对上述截取图像进行边缘检测形成的伪边缘,保留托辊的真实边缘;利用形态学闭运算连接托辊边缘缝隙,通过边界跟踪获得托辊闭合轮廓并进行种子区域填充,实现托辊自动分割;最后根据所得托辊二值图像闭合轮廓,基于轮廓像素点遍历在原红外图像完成托辊的自动定位。实验表明,本方法可快速实现托辊的自动分割和定位,为带式输送机托辊的运行状态监测奠定了基础。
[Abstract]:Due to the lack of effective monitoring of the key parts of belt conveyor in coal mine, the traditional inspection and temperature monitoring methods have many problems, such as heavy workload and many blind spots. In this paper, an infrared image segmentation and location algorithm of belt conveyor roller based on connected component is proposed, and the infrared image of belt conveyor collected by the inspection robot along the track is processed. The vertical and horizontal projection is used to intercept the area of the roller to reduce the influence of the bracket, conveyor belt and background on the subsequent image processing, and the image edge is filtered based on the information of the long-axis ratio and the area of the connected component. Eliminating the pseudo-edge formed by edge detection of the intercepted image, preserving the real edge of the roller, connecting the edge slot of the roller by morphological closed operation, obtaining the closed contour of the roller by boundary tracking and filling the seed area. Finally, according to the obtained binary image of the roller closed contour, based on the contour pixel traversal in the original infrared image to complete the automatic positioning of the roller. The experimental results show that the method can quickly realize the automatic partition and positioning of the rollers and lay a foundation for monitoring the running status of the rollers of belt conveyors.
【作者单位】：西安科技大学机械工程学院;
【基金】：国家自然科学基金(U1361121)
【分类号】：TD528.1;TP391.41

【相似文献】