Automatic bacterial colony screening method based on bacterial colony morphological characteristics

Abstract

The invention discloses an automatic bacterial colony screening method based on bacterial colony morphological characteristics. The method includes the steps that firstly, a bacterial colony image is input; secondly, a conversion model of a bacterial colony image pixel coordinate system and a world space coordinate system is established; thirdly, the acquired bacterial colony image is preprocessed, and the preprocessed bacterial colony image is acquired; fourthly, the preprocessed bacterial colony image is subjected to binarization processing, a binary bacterial colony image is obtained and is segmented according to the communication principle, and a binary bacterial colony area is obtained after segmentation; fifthly, the morphological characteristics of the binary bacterial colony area are calculated; sixthly, bacterial colonies obtained in the fifth step are screened. According to the method, the bacterial colonies are screened according to the morphological characteristics of the individual bacterial colonies and are compared with a morphological characteristic sample base of excellent bacterial colonies so that the screening accuracy of the bacterial colonies can be guaranteed, universality is high, and meanwhile, bacterial colony screening accuracy and precision of the method are proved through a large number of bacterial colony screening experiments performed on a bacterial colony selector.

Description

Technical field [0001] The invention relates to the field of biomedical image processing, in particular to a colony automatic screening method based on colony morphological characteristics. Background technique [0002] Colony culture is a basic task for quality inspection in agriculture, food, medical and health analysis and the production of antibiotics, and colony screening and statistics are the most important links. Colony screening can select high-quality and high-yielding strains for pharmaceuticals. The experience of colony screening shows that the quality of the colony has a certain relationship with its biological characteristics, and the colony morphology is one of the important biological characteristics. In the past, most of the work of colony screening and statistics was that the staff observed the petri dishes with the naked eye to obtain the approximate colony size, shape and color, and then classified, screened and counted the colonies. This approach requires h...

AI Technical Summary

Problems solved by technology

This approach has high professional requirements for staff, tedious work, strong subjectivity, large errors, and poor reproducibility

In addition, the amount of test samples used for general colony screening is often relatively large. The traditional manual screening method has long cycles, low efficiency, and complicated procedures. A more effective screening method is needed.

[0003] Most of the existing methods use the global threshold method as the colony identification and screening method, which still has the disadvantages of long image processing time, low identification rate of colonies with similar color to the culture medium, and high false screening rate

Images