|
|
基于改进Zernike矩的圆形标志高精度定位方法
|
Abstract:
在基于视觉的户外边坡监测中，针对圆形标志中心定位精度低，实效性差的难题，提出一种改进的Zernike矩圆形标志高精度定位方法。首先对圆形标志图像进行二值化和形态学滤波处理，然后利用Canny算子对二值化图像进行边缘检测，实现圆形标志边缘的粗定位。其次结合Zernike矩法和大津法，对粗定位得到的圆形标志边缘进行重定位，通过Zernike矩法来重新定位边缘的位置，并基于改进的大津法来确定圆形标志的亚像素边缘点。同时，利用基于灰度梯度优化的方法来对亚像素边缘进行细化，进一步提高定位的精度。最后利用最小二乘椭圆拟合法来实现圆形标志中心的高精度定位。通过实验证明，改进的Zernike矩圆形标志高精度定位方法能够有效提高圆形标志的定位精度和实效性，并具有更高的运行效率。
In vision-based outdoor slope monitoring, an improved high-precision positioning method for circular signs with Zernike moments is proposed to address the challenges of low accuracy and poor effectiveness of circular sign center positioning. Firstly, the circular sign image is binarized and morphology filtered, and then the edge detection of the binarized image is carried out using Canny operator to achieve the coarse positioning of the edge of the circular sign. Secondly, the edge of the circular logo obtained by coarse localization is relocated by combining Zernike moment method and Otsu method to relocate the position of the edge and determine the sub-pixel edge points of the circular logo based on the improved Otsu method. Meanwhile, a grey scale gradient optimization based method is used to refine the sub-pixel edges to further improve the accuracy of the localization. Finally, the least-squares ellipse fitting method is used to achieve high-precision localization of the center of the circular sign. It is proved through experiments that the improved Zernike moment circular sign high-precision localization method can effectively improve the localization accuracy and effectiveness of circular signs with higher operational efficiency.
| [1] | Zhou, T. and Papanikolopoulos, N. (2011) Enhancing the Randomized Hough Transform with K-Means Clustering to Detect Mutually-Occluded Ellipses. 2011 19th Mediterranean Conference on Control & Automation (MED), Corfu, 20-23 June 2011, 2011, 327-332. https://doi.org/10.1109/MED.2011.5983040 |
| [2] | 邹荣, 赵稼宸, 凌俊, 等. 基于Hough投票空间的椭圆图像特征亚像素提取方法[J]. 光学技术, 2016, 42(2): 141-145. |
| [3] | 王林波, 王延杰, 邸男, 等. 基于几何特征的圆形标志点亚像素中心定位[J]. 液晶与显示, 2014, 29(6): 1003-1009. |
| [4] | 陈向阳. 基于LabVIEW的钢拱架定位系统开发[D]: [硕士学位论文]. 武汉: 华中科技大学, 2021. |
| [5] | 张萍萍, 李童, 李茹, 等. 一种改进的Sobel图像边缘检测算法及其实现[J]. 电视技术, 2022, 46(5): 42-45. |
| [6] | 焦瑾瑾. 基于Roberts边缘检测的数字媒体激光重现方法[J]. 激光杂志, 2022, 43(5): 197-200. |
| [7] | 姚智超, 楚晓亮, 范筠益, 等. 基于Prewitt算子的X波段雷达有效波高反演研究[J]. 系统工程与电子技术, 2022, 44(4): 1182-1187. |
| [8] | 陈思吉, 王晓红, 李运川. 改进Laplace的无人机图像边缘检测算法研究[J]. 测绘工程, 2021, 30(2): 36-44. |
| [9] | Guo, L.G. and Wu, S.T. (2023) FPGA Implementation of a Real-Time Edge Detection System Based on an Improved Canny Algorithm. Applied Sciences, 13, 870. https://doi.org/10.3390/app13020870 |
| [10] | 刘利平, 孙建, 孙文悦. 改进Zernike矩的亚像素圆孔类零件测量方法[J]. 电子测量技术, 2023, 46(4): 69-77. |
| [11] | 张炳星, 高军伟, 王建冲, 等. 基于机器视觉的圆形垫圈尺寸测量系统设计[J]. 工具技术, 2023, 57(7): 141-145. |
| [12] | 韩东, 李煜祺, 武彦辉. 基于高斯拟合的亚像素边缘检测算法[J]. 计算机应用与软件, 2018, 35(6): 210-213+229. |
| [13] | 陈绍杰, 郑守住, 赵文明, 等. 视频测量椭圆标志中心点的高精度定位方法[J]. 测绘科学, 2018, 43(7): 121-127. |
| [14] | 刘悦, 朱均超. 基于改进Zernike矩的光斑图像亚像素边缘检测算法[J]. 激光杂志, 2021, 42(5): 32-35. |
| [15] | 郑守住, 汪本康, 陈绍杰, 等. 高速影像中圆形标志点的自动识别与定位[J]. 测绘科学技术学报, 2020, 37(4): 380-387. |
| [16] | Ghosal, S. and Mehrotra, R. (1993) Orthogonal Moment Operators for Subpixel Edgedetection. Pattern Recognition, 26, 295-306. https://doi.org/10.1016/0031-3203(93)90038-X |
| [17] | 舒启林, 山博. 改进canny算子的亚像素定位算法[J]. 机械设计与制造, 2018(10): 165-168. |
| [18] | 刘军, 赵剑英, 池云. Zernike矩算法实现焊缝中心线亚像素检测[J]. 沈阳理工大学学报, 2021, 40(4): 1-5, 10. |
| [19] | 高世一, 赵明扬, 张雷, 等. 基于Zernike正交矩的图像亚像素边缘检测算法改进[J]. 自动化学报, 2008, 34(9): 1163-1168. |
| [20] | 刘光宇, 王帅, 曹禹, 等. 几种经典阈值分割方法在图像处理中的应用研究[J]. 黄冈职业技术学院学报, 2022, 24(4): 99-103. |
| [21] | 王靖, 魏亮, 向文豪, 等. 考虑圆形特征边缘模糊和偏心误差修正的高精度相机标定方法[J]. 红外与激光工程, 2021, 50(12): 369-379. |
