基于改进Hough圆检测算法的明场微滴图像识别

Bright Field Droplet Image Recognition Based on Improved Hough Circle Detection Algorithm

针对数字聚合酶链式反应(PCR)明场微滴图像的微滴数目统计和直径计算方法需借助人工处理费时费力,且识别精度不高的问题,提出一种改进Hough变换的快速又准确的微滴检测算法。首先,利用边缘点的梯度信息和圆的几何性质确定候选圆;其次,若设定范围内的边缘点到候选圆圆心的距离与半径近似相等,则将此圆判定为真实圆;最后,将近邻的真实圆中半径最大的圆作为最终的检测结果。经对比实验验证,文中算法能明显提高微滴识别的效率和直径测量的精度。

Aiming at the number statistics and diameter calculation method of droplets in digital polymerase chain reaction(PCR)bright field droplet image need the help of manual processing,which is time-consuming and laborious,and the recognition accuracy is not high,a fast and accurate droplet detection algorithm based on improved Hough transform is proposed.Firstly,the candidate circle is determined by using the gradient infor-mation of edge points and the geometric properties of the circle;Secondly,if the distance between the edge point within the set range and the center of the candidate circle is approximately equal to the radius,the circle is determined as a real circle;Finally,the circle with the largest radius in the nearest real circle is taken as the final detection result.Comparative experiments show that the algorithm can significantly improve the efficiency of droplet recognition and the accuracy of diameter measurement.