Research on Rapid Detection of Total Bacteria in Juice Based on Biomimetic Pattern Recognition and Machine Vision

摘要

In order to develop an automatic and rapid detection method for enumeration of total bacteria in juice, biomimetic pattern recognition and machine vision were employed. The characteristic data, such as shape, texture and color features, were acquired by using the machine vision technology from bacteria images in varieties of juice. Based on multi-weight higher order neuron network, the recognition models were established which can achieve the imitation of human learning, memorizing and judging. By applying the principle of statistics, the detection results of new method showed no difference, compared with the traditional method in apple juice, tomato juice and carrot juice. The new method simplifies experimental preparation and shortens judgment time, especially in sample test on the spot and monitoring production site. Moreover, by using this rapid detection method, total bacteria counts in samples could be accurately enumerated within 1 h, which was much less than 24-48 h by using the traditional method.