Dragon fruit is a tropical fruit, which favored in many countries because of its rich nutrition and medicinal value. In Vietnam, dragon fruit is a fruit with a high percentage of exports, and they have been evaluated and sorted according to the standard of importing countries. The workshops for evaluating and sorting dragon fruit are based on human workers, they use the hands to rotate the fruit to determine the area of defective areas qualitatively and use basic scales to weigh dragon fruits that cause errors of grading. Therefore, this study proposes an automatic dragon fruit sorting system to replace humans in sorting to reduce costs and increase accuracy for grading. The automatic sorting system has a yield of approximately 3 tons/h for a module, the results have the system's experiments of much more than 96%. Dataset is collected from dragon fruit farms, packaging facilities. The image of the actual system is represented in the 'system.rar' file.