Fruit color is one of the important indicators of quality and commodities. Three kinds of the traditional methods are used to evaluate fruit color, including artificial visual identification, fruit standard color cards and color measurement instrument. These methods are needed to be conducted in the field by persons, which are time-consuming and labored, and also difficult to obtain the dynamic color information of the target fruits in the growth process. This study developed a practical technology for automatically monitoring the fruit color in the whole progress based on machine vision, the apple fruits were photographed by infrared camera, and the dynamic color change information were obtained through applying MATLAB software to process the image. Color change rule of the whole apple and fixed local position apple was analyzed. The results showed that the trend of the fixed local position R/G ratio was similar to that of the whole fruits. Therefore, the fixed local position color information was representative for whole apple in the corresponding period. The R/G ratio of apple was close to 1 when the bag was just removed, with the bag-removed days increased, the apple’s color became more red, and R/G ratio was increased gradually, the R/G ratio was about 1.4 on October 25th, and the fruit was bright red. We can get dynamic color information of apple fruit through automatic monitoring technology based on machine visions. This technology can lay the foundation for analyzing the influence factors of apple color and building color model, and also providing an important reference for picking and presorting apple in suitable period.