p. 289-292
This paper presents a computer simulation study of natural wolfberry recognition based on color feature. We performed a computer simulation of the recognition between natural and sulfur-smoked wolfberries by using color classification for food safety. First, two types of images of the natural and sulfur-smoked wolfberries were collected in the Internet. These images were converted from RGB to CIELab color space and the color feature vectors were obtained by the combination of chromatic components. Then the color vectors of these two types of images were classified by the k-means clustering algorithm in the training process. By means of the color classification, the recognition between natural and sulfur-smoked wolfberries is performed. Simulation results showed that the proposed method provides a recognition rate of about 80% in the collected image dataset.