Open AccessAn Automated Method for Bacterial Flora Counting Based on Image Analysis
Author(s) -
Xinyang Wu,
Jinke Liu,
Xingkai Zheng,
Xi Chen,
Yu-Tao Que,
Borong Ma
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1673/1/012016
Subject(s) - computer science , bacterial colony , hough transform , filter (signal processing) , artificial intelligence , image (mathematics) , computer vision , workload , flora (microbiology) , pattern recognition (psychology) , data mining , biology , genetics , bacteria , operating system
How to quickly detect the characteristics of water quality flora, such as bacterial flora number, is a key problem of water quality safety, the traditional manual counting method has some serious defects such as low efficiency, low accuracy, easy to cause visual fatigue and hard to deal with large samples. This paper designs an automated method for bacterial flora counting based on image analysis, including such modules as pretreatment of bacterial image classification and colony counting of culture medium images by filter method (using Hough circle detection algorithm and a circle deduplication algorithm). The accuracy of experimental results is above 90%, which is in line with international standards, and the time complexity of the algorithms is up to the order of square. This method can solve the problems of heavy workload and low efficiency in traditional manual counting method.