摘要
针对传统Hough变换进行圆检测,计算量过大、检测同心圆精度不高、自动化程度低等缺点,提出一种基于连通区域标记算法的圆检测算法。该算法首先通过连通区域标记算法对图像进行处理得到一个圆,解决了传统Hough变换计算量过大的问题,再根据圆的特性确定其圆心及半径,从而避免了检测同心圆精度不高的问题。最后,分别取圆心的8邻域像素为圆心做圆,找到最优圆并将其与检测得出的圆进行比较来确定最终的圆,以达到自动化的目的。实验结果表明,提出的算法可以正确地检测出圆并具有很高的检测精度同时比Hough变换计算量小、自动化程度较高。
In order to overcome the shortcomings of traditional Hough transform in the circle detection, such as excessive computation, low precision of detecting concentric circles and low degree of automation, a circle detection algorithm based on connected region labeling algorithm is proposed. This new algorithm firstly solves the problem of excessive calculation which the traditional algorithm to process the image for the circle. Secondly, confirming the center and radius of the circle according to the characteristics, it avoids the detection problem of concentric circles. Finally, for the aim of automation, the 8 neighborhood pixels of the center of the circle are taken to determine the best and final circle. The experimental results show that the proposed algorithm can detect the circle accurately. Meanwhile, it's less computational and more automated than the Hough transform.
引文
[1]郭斯羽,翟文娟,唐求,等.结合Hough变换与改进最小二乘法的直线检测[J].计算机科学,2012,39(4):196-200.
[2]秦开怀,王海颍,郑辑涛.一种基于Hough变换的圆和矩形的快速检测方法[J].中国图象图形学报,2010,15(1):109-115.
[3]毛顺兵.Hough变换和轮廓匹配相结合的瞳孔精确检测算法[J].计算机应用,2016,36(5):1415-1420.
[4]应俊,李强.基于改进Hough变换的电梯轿厢内人数统计算法[J].计算机工程与应用,2014,50(19):165-168.
[5]黄永林,叶玉堂,陈镇龙,等.一种新的快速Hough变换圆检测方法[J].电子测量与仪器学报,2010,24(9):837-841.
[6]王凯,施隆照.基于FPGA的快速连通区域标记算法的设计与实现[J].计算机工程与应用,2016,52(18):192-198.
[7]施隆照,王凯.基于连通区域的复杂车牌的字符分割算法[J].计算机工程与设计,2016,37(8):2138-2142.
[8]张恒,倪永婧.面向目标特征提取的连通区域标记算法[J].计算机与网络,2015(7):58-61.
[9]Hough P V C.Method and means of recognizing complex patterns:US3069645[P].1962.
[10]Illingw J,Kitter J.A survey of the Hough transforms[J].Computvision Grapics Image Process,1988,44(1):87-116.
[11]张铮,王艳平,薛桂香.数字图像处理与机器视觉[M].北京:人民邮电出版社,2010.
[12]Gonzalez R C,Woods R E.数字图像处理[M].2版.北京:电子工业出版社,2013.
[13]Rosenfeld A,Pfaltz L.Sequential operations in digital picture processing[J].Journal of the ACM,1966,13(4):471-494.
[14]冯海文,牛连强,刘晓明.高效的一遍扫描式连通区域标记算法[J].计算机工程与应用,2014,50(23):31-35.
[15]马爽,姜月秋,赵阳,等.一种基于连通标记的教室人数检测算法研究[J].沈阳理工大学学报,2016,35(1):17-21.
[16]Hernandez-Belmonte U H,Ayala-Ramirez V,SanchezYanez R E.A comparative review of two-pass connected component labeling algorithms[J].Advances in Soft Computing,2011,7095(12):452-462.