Application of SVM in the food bacteria image recognition and count

In order to overcome the time-consuming and difficulties in the bacterial recognition and counting during the traditional process of manual detection of food contamination, bacteria staining technology, microscopic image processing, and support vector machines (SVM) are applied to realize the rapid detection. According to the characteristics of microscopic image, we study the preprocessing, binary processing, feature extraction, bacterial recognition and counting in this paper. Compared with the results recognized by human eye, SVM can effectively distinguish the bacteria from non-bacteria in the image, and greatly reduce the detection time of each sample. A new bacterial count method is proposed based on the results of SVM, and difference between the result of the new method and manual counting is little.