Since the “14th Five-Year Plan”, with the continuous deepening of the “three rural work”, the quality inspection and grading of potatoes have played a decisive role in the expansion of the potato industry and the development of the processing industry. Aiming at the problems of traditional manual potato quality classification, such as high labor intensity, low efficiency, easy to cause secondary damage, and inability to judge internal quality, this article elaborated on the research progress of domestic and foreign hyperspectral detection technology, near-infrared spectroscopy detection technology, machine vision detection technology, and ultrasonic detection technology in the application of internal and external quality detection of potatoes. Through research and analysis, the problems facing non-destructive detection technology for potato quality were proposed, and prospects for the selection of imaging systems, multimodal detection technology, algorithm optimization, and intelligent detection technology were discussed, which could provide a reference for in-depth research and practical application of non-destructive detection technology for potato quality.