The invention discloses a rapid apple variety identification method based on hyperspectral imaging. The method comprises the following steps: obtaining an average reflection spectral value of three varieties of samples at a near-infrared full-wave band by adopting a near-infrared hyperspectral technology by virtue of a series of image processing, establishing a mathematic model of a K nearest algorithm by virtue of a Metalab function and an operational command, calculating by using the model, so that an apple variety can be accurately identified, and the highest identification accuracy by using the K nearest algorithm can reach 100 percent. The method is a novel method which can more accurately, more conveniently and rapidly identify the apple variety; and by acquiring the different near-infrared hyperspectral spectrum of various varieties, the individual data of the apple variety can be enriched and used as evidence for fingerprint identification and mutation identification.