In order to develop an automatic and rapid detection system for enumeration of total viable bacteria in vegetables, viable staining technology, computer vision, image processing and artificial neural network were used. The methylene blue was used as a vital stain agent to distinguish the viable bacteria from the non-viable bacteria, and the CCD digital camera with 5 200 000 pixels was used to capture the staining effect within the bacteria. Moreover, a new algorithm which could extract the viable bacterial images successfully was determined. According to the bacterial shape features, eight feature parameters (eccentricity, circularity, rectangle degree, et al.) were selected to be the input vectors of the artificial neural network in order to identify the bacteria. The proposed detection system was easy to operate. By using this rapid detection system, total viable bacteria counts in samples could be accurately enumerated within 40 min, which was much less than 48 h by using the traditional aerobic plate count method. Moreover, comparisons of detective results of total bacteria counts by rapid automatic detection system and aerobic plate count method were made, they were closely correlated (R2=0.9987). And the T test results also showed that there was no significant difference (P＞0.05) between these two detection results. Therefore, the rapid detection system will greatly adapt to the requests of on-site rapid detection technique for the safety of agricultural products.