Research On Rapid Detection Of Bacillus Cereus In Food By Computer Vision

As food-borne pathogen increasingly involved in food-poisoning outbreaks, there is an interest to study on the rapid detection methods of food microbe, in order to improve the deficiency of traditional methods and meet the growing demand for fast and accurate detection methods of pathogenic bacteria to ensure the food quality and safety. Bacillus cereus is one of the common food-borne pathogens that can cause food poisoning, and widely distributed in the environment because of their resistant spores.At Present, detection and identification methods for B. cereus are still based on traditional culture-based, which is always a complicated and time-consuming process and difficult to meet real-time monitoring of food quality and safety requirements. Though many rapid detection methods are used to detection B. cereus, they are always demanded large-scale equipment. There is necessary to establish a simple, rapid and accurate detection method of B. cereus.The aim of this study was to develop a rapid detection method of B. cereus by computer vision, and establish a rapid detection system. Based on the rapid detection system, which our own laboratory are developing, the paper mainly researches both the method of pre-treatment about how to treat the food samples and collected B. cereus, and microscopic image analysis for quantitative measurement and feature identification of the specificity in B. cereus. And a rapid detection system was established to detect B. cereus in food, and preliminarily evaluated its application.In this thesis, the main contents and results are as follows:1. Make basic research on B. cereus spores physical characteristics and carry out the preliminary screening of B. cereus in food, because of its heat resistance. By comparing the effect of different treatment methods and conditions on spore germination, the optimum germination factors were determined. On this condition, there is the highest germination rate and minimum inactivation rate of spores.2. Study on the cell staining methods of B. cereus. After compared the dyeing effect of different chromogenic reagents and cell staining methods, selected the 5-bromo-4-chloro-3-indolyl-glucopyranoside as the best chromogenic substrate, which is cleaved by enzyme and liberate the specific chromophor. And it has the advantages of faster coloring and fewer impurities. By comparing the staining effect of the adding order and dosage of chromogenic reagent, staining B. cereus after a short-term fermentation will get a good result, which will be beneficial to observe cell morphology and count;3. The digital image of B. cereus are analyzed and processed by microscopic image, according to threshold segmentation by color characteristics and mathematical morphology. After image analysis and processing, established a set of pattern to obtain and analysis B. cereus sample images, which selecting eight effective characteristics parameters, including shape and color features. After researching the cell characteristics, also established a set of pattern to identify B. cereus on the basis of BP networks, and experiments showed that the system can effectively identify and count target image;4. Comparison of the traditional culture methods and the rapid detection system, which is developed by our lab, the results show that the rapid detection method is more rapid than the traditional culture method, and the detection time it takes is only about 5h. There is a linear relationship with a good correlation (R2>0.99) and no significant difference (P>0.05) between two methods. And the rapid detection system is more sensitivity, specificity and practicality after preliminarily analyzing the detection limits of the system.The thesis mainly research on the rapid detection method of B. cereus based on computer vision and specific staining method. This paper not only put forward new ideas on rapid detection of B. cereus in food, but also expanded the application of microbiological rapid detection system and provided theoretical basis and technical support for the perfectation of microbial rapid detection system.