Abstract: With increase of global population and high demand for fruit and vegetable quality, efficiency and accuracy of fruit and vegetable quality inspection have become focal points in food industry．Traditional methods of fruit and vegetable quality inspection face challenges such as high labor intensity, low efficiency, and difficulty in ensuring accuracy．Application of deep learning in computer vision tasks of fruit and vegetable quality inspection was described in detail．Deep learning could achieve efficient and accurate inspection by learning and simulating human visual recognition process to detect quality of fruits and vegetables．Specific applications include chromatic analysis, shape analysis, texture analysis, size analysis, disease identification, and ripeness recognition．However, deep learning in fruit and vegetable quality inspection still faces challenges in data collection, model training complexity, and achieving real-time performance．Future research could focus on improving inspection accuracy, processing speed, and reducing costs, exploring integration of deep learning with internet of things, big data and other technologies, to achieve more intelligent and automated fruit and vegetable inspection system．Application of deep learning in fruit and vegetable inspection has a broad development prospect and will bring a significant impact on fruit and vegetable production and consumers．