针对现有的直升机桨叶欠曝光图像中圆形标记点检测方法存在自适应能力不强、速度慢、精度不高的问题，提出了基于YOLOv3（you only look once）与分水岭的直升机桨叶欠曝光图像圆形标记点检测方法.首先，将采集的真实桨叶欠曝光图像中的圆形标记点进行标注后，制作成数据集，并训练YOLOv3网络；其次，用训练好的YOLOv3网络检测出圆形标记点区域；再次，改进传统分水岭标记提取方式，采用多线程技术并行在各圆形标记点区域内进行分水岭变换，得到圆形标记点边缘检测结果；最后，采用最小二乘圆拟合和奇异点去除法实现圆形标记点的精确定位.研究者通过对多幅欠曝光桨叶图像中圆形标记点进行检测实验，验证了该方法具有自适应能力强、速度快、精度高的优点，并已将其用于直升机桨叶欠曝光图像圆形标记点的检测.
A method of circular marker detection in under-exposed helicopter blade images based on YOLOv3 (you only look once) and watershed algorithm is proposed, aiming to improve detection adaptability, speed up the detection and obtain accurate position of circular markers. Firstly, the real under-exposed blade images are labeled for dataset making, on which the YOLOv3 network is trained. Secondly, the circular marker regions in blade images from testing dataset are detected by the trained YOLOv3 network. Thirdly, the traditional watershed marker detection method is improved, and the circular marker regions are used for the watershed transformation by the multi-threading technology in parallel, and the edge detection results of the circular marker are obtained. Finally, the circular markers are accurately located by the least square circle fitting and the method of removing the singular points. The proposed method is proved to be adaptable, fast and accurate by a number of experiments in many under-exposed helicopter blade images, and has been applied to circular marker detection in under-exposed helicopter blade images.
[1] 杨景豪, 刘巍, 刘阳, 等. 双目立体视觉测量系统的标定[J]. 光学精密工程, 2016, 24(2):300-308. Yang J H, Liu W, Liu Y, et al. Calibration of binocular vision measurement system[J]. Optics and Precision Engineering, 2016, 24(2):300-308. (in Chinese)
[2] 欧巧凤, 赵平均, 熊邦书, 等. 基于立体视觉的旋翼共锥度动态测量系统精度分析[J]. 仪器仪表学报, 2015, 36(8):1692-1698. Ou Q F, Zhao P J, Xiong B S, et al. Accuracy analysis of the measuring instrument for taper angle of running rotor blades based on stereo vision[J]. Chinese Journal of Scientific Instrument, 2015, 36(8):1692-1698. (in Chinese)
[3] 毛东月, 谢正祥, 贺向前, 等. 自适应双向保带宽对数变换及低照度图像增强[J]. 中国图象图形学报, 2017, 22(10):1356-1363. Mao D Y, Xie Z X, He X Q, et al. Adaptive bilateral logarithm transformation with bandwidth preserving and low-illumination image enhancement[J]. Journal of Image and Graphics. 2017, 22(10):1356-1363. (in Chinese)
[4] 周封, 杨超, 王晨光, 等. 基于随机Hough变换的复杂条件下圆检测与数目辨识[J]. 仪器仪表学报, 2013, 34(3):622-628. Zhou F, Yang C, Wang C G, et al. Circle detection and its number identification in complex condition based on random Hough transform[J]. Chinese Journal of Scientific Instrument, 2013, 34(3):622-628. (in Chinese)
[5] 叶峰, 陈灿杰, 赖乙宗, 等. 基于有序Hough变换的快速圆检测算法[J]. 光学精密工程, 2014, 22(4):1105-1111. Ye F, Chen C J, Lai Y Z, et al. Fast circle detection algorithm using sequenced Hough transform[J]. Optics and Precision Engineering, 2014, 22(4):1105-1111. (in Chinese)
[6] Akinlar C, Topal C. EDCircles:a real-time circle detector with a false detection control[J]. Pattern Recognition, 2013, 46(3):725-740.
[7] 王娅. 血液红细胞图像自适应标记分水岭分割方法[J]. 中国图象图形学报, 2017, 22(12):1779-1787. Wang Y. Adaptive marked watershed segmentation algorithm for red blood cell images[J]. Journal of Image and Graphics, 2017, 22(12):1779-1787. (in Chinese)
[8] Redmon J, Divvala S, Girshick R, et al. You only look once:unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016:779-788.
[9] Ren S, He K, Girshick R, et al. Faster R-CNN:towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems, 2015:91-99.
[10] Liu W, Anguelov D, Erhan D, et al. SSD:single shot multibox detector[C]//European Conference on Computer Vision. Cham:Springer, 2016:21-37.
[11] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016:770-778.
[12] Erisoglu M, Calis N, Sakallioglu S. A new algorithm for initial cluster centers in k-means algorithm[J]. Pattern Recognition Letters, 2011, 32(14):1701-1705.
[13] Ahmed F, Tarlow D, Batra D. Optimizing expected intersection-over-union with candidateconstrained CRFs[C]//Proceedings of the IEEE International Conference on Computer Vision, 2015:1850-1858.
[14] 王宇, 陈殿仁, 沈美丽, 等. 基于形态学梯度重构和标记提取的分水岭图像分割[J]. 中国图象图形学报, 2008, 13(11):2176-2180. Wang Y, Chen D R, Shen M L, et al. Watershed segmentation based on morphological gradient reconstruction and marker extraction[J]. Journal of Image and Graphics, 2008, 13(11):2176-2180. (in Chinese)
[15] Koyuncu C F, Akhan E, Ersahin T, et al. Iterative H-minima-based marker-controlled watershed for cell nucleus segmentation[J]. Cytometry Part A, 2016, 89(4):338-349.
