基于ICM的精密指针式仪表自动读数方法

针对精密指针式仪表自动读数算法自适应性较差的问题,提出一种基于ICM的精密指针式仪表自动读数方法。该方法基于二值化后仪表图像的噪声信息分布特性,利用ICM点火时序矩阵在图像相邻像素间的并行递进关系,研究和实现图像的腐蚀和膨胀,并设计加权最大类间方差法自适应地计算迭代次数,消除刻度周边噪声,提取到刻度圆弧所在位置。最后根据圆弧极径、极角等信息对刻度区域做极坐标变换输出图像,采用距离法对极坐标变换后图像进行读数判读。实验表明,自动读数结果与实际数值相比误差均小于0.2%,算法稳定可靠。

Auto reading method for precision pointer meter based on ICM

Aiming at the problem of adaptivity in traditional auto reading method for precision pointer meter, an auto reading method based on intersect cortical model(ICM) is proposed. The proposed approach designs ignition erosion and dilation by combining the parallel progressive relationship of ICM sequential matrix among the neighbor pixels and the distribution of noise information in binary meter image, and then unites weighted Otsu to calculate adaptively ICM iterative times to eliminate noise around the scale. By using the proposed method, the arc position of the scale is accurately extracted. Finally, polar translation image is derived by using information of arc, and the meter reading is calculated by the relative position between scale and index. Experimental results show that the proposed method can meet the applications requirements and the relative error between meter auto readings and actual readings is less than 0.2%.