An automatic control system based on computer vision and fuzzy logic methods was developed to control the performance of a rice whitening machine. By monitoring the performance of the whitening machine using the developed ACS, it was found that only at the first stage of monitoring, there was a significant difference between ACS and human operator in terms of the pressure level. So, the difference was at one level: “High” selected by ACS compared with “Very High” chosen by the human operator. This difference could be justified by overlapping the boundary ranges of fuzzy MFs, where the expert, based on his empirical and uncertain knowledge, chooses different crisps to judge the output grain quality. It was revealed that on the average, the ACS was 31.3% faster than the human operator. The levels of the qualitative indices of the sampled products at each stage of the control process were approximately close. This implied that the performance monitoring of the rice whitening machine resulted in an almost uniform rice product during the control process. One of the advantages of the developed system was its setup flexibility, so that the overall adjustments of the control program (such as number of sampling, sampling interval, number of images per sample) and even the structure of the fuzzy inference unit could be altered according to the preference of each rice mill operator. Evaluation of the samples obtained from the discharge section of the whitening machine at different stages of the control process showed that the decisions made by the developed ACS led to a satisfactory improvement in the quality of the output product.