Pointer Meter Reading Algorithm Based on Improved GA-Otsu and RANSAC Regression
In order to solve the problems of low manual reading efficiency and accuracy of pointer instruments,a pointer instrument reading method based on improved GA-Otsu and RANSAC regression is proposed.ABF is used to filter the texture and noise of the pointer instrument image,and the dial image is extracted by Hough gradient method and Mask'mask method.The separated pointer region is obtained by image segmentation algorithm based on improved GA-Otsu,and the pointer thinning map is extracted by morphological processing.RANSAC algorithm is used to fit the straight line where the pointer center is located.The angle value is calculated,and the instrument reading recognition is completed by combining the range information and the angle method.The experimental results show that the proposed algorithm can effectively separate the pointer target from the background.Compared with the original algorithm,the recognition speed is improved by 42.34％,the average relative error of recognition is less than 1.15％,and it is robust against different illumination and shadow interference.
NETL
NSTL
万方数据
