基于图像处理检测抑菌圈边缘的研究  

Study on Image-Processing-Based Edge Detection of Bacteriostatic Ring

作　　者：张全法[1] 杨莉[1] 张洋[1] 

机构地区：[1]郑州大学物理工程学院,河南郑州450052

出　　处：《河南农业大学学报》2008年第1期104-107,共4页Journal of Henan Agricultural University

摘　　要：提出了一种自动检测抑菌圈边缘的新方法.在对灰度图像进行中值滤波及锐化处理后,用灰度阈值二值化为黑白图像.然后用圆环形模板在黑白图像上扫描,确定抑菌圈中心的位置.在该位置附近,通过对半径直方图数据的处理,消除抑菌圈内部菌落的影响,用曲线拟合,可自动地检测到抑菌圈的边缘.对于3种不同的图像获取设备,成功率分别为100%,100%,91%.A new method was presented for auto-edge-detection of bacteriostatic ring. After mean value filtering and enhancement were conducted, gray images were translated into binary images. Then a template of ring shape was used to locate the center of each bacteriostatic ring by moving the template through the binary image. Around each center that was found, radius histogram data were used to eliminate the influence of bacterial colony that dispersed in each bacteriostatic ring. Finally, curve-fitting was used to detect the edge of each bacteriostatic ring automatically. For the three differenct image equipments, the success ratio was 100% , 100% and 91% ,respectively.

关 键 词：抑菌圈 边缘检测 半径直方图 图像处理 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]