Abstract: In order to explore the genetic diversity of apricot leaf morphology, and to facilitate the collection of apricot germplasm resources and variety identification. In this study, the comparative analyses were carried out by using various methods, such as descriptive statistics, principal component analysis and cluster analysis. The results showed the coefficients of variation (cv) of the 13 graded traits were all above 20%, and the three leaf traits of leaf tip length, leaf margin serration depth, and position of the widest part of the leaf blade differed greatly, which can be used as traits to distinguish leaf morphology of different apricot varieties. The magnitude of variation for the 15 quantitative traits ranged from 4.16% to 29.5%, which showed that the degree of leaf variation among apricot varieties was high, and there were a variety of types and rich genetic diversity among apricot varietal resources. Through correlation analysis, it was found that there was a highly significant positive correlation between four traits: leaf blade roundness, leaf blade rectangularity, leaf apex angle and leaf base angle; leaf base shape was highly significantly and positively correlated with leaf base angle with a correlation coefficient of 0.92, and highly significant negatively correlated with leaf length-to-width ratio; and leaf roundness was significantly negatively correlated with leaf tip length, which indicated that tip length of leaf to a certain extent could reflect the leaf shape. Through principal component analysis, the 15 quantitative traits were divided into four composite factors, and the cumulative contribution rate reached more than 85.04%. The results showed that the first principal component consisted of five traits, including leaf area, leaf circumference, leaf length, leaf width, and petiole length, which mainly reflected the traits related to the size of the leaf blade of the apricot germplasm resources; the second principal component reflected the traits related to the color of the leaf blade; and the third and the fourth principal components reflected the traits related to leaf blade shape and leaf blade serration ratio, respectively. The four dominant factors of leaf size, leaf color, leaf shape and average serration height can be used as the main phenotypic indexes for classification of apricot germplasm resources. Apricot germplasm resources were classified into 8 taxa based on cluster analysis. Taxon Ⅰ consisted of only 2 germplasm, Zao Dahuang and Chaoxianxing, which were in the large-leaved-cuneate category; Taxon Ⅱ and Ⅲ both consisted of 1 material, which was Huanna and Zhuolumuguxing, respectively; Taxon Ⅳ consisted of 18 germplasm, which were in the small-leaved-deep-green category; Taxon Ⅴ consisted of 1 Central Asian germplasm, which was in the Stella category; Taxon Ⅵ consisted of 1 material, which was in the large-leaved-cardinal category; Taxon VII consisted of 12 germplasm, which were in the large-leaved-truncate category; Taxon VIII was further classified into 6 subgroups. Taxon VII consists of 12 germplasm, which is large-leaved-truncate; Taxon VIII is further divided into six subgroups. In this study, based on the application of image recognition technology, we were able to rapidly evaluate the leaf morphology of apricot varieties and resources, and increase the length of the leaf tip, the depth of the leaf margin serration, and the position of the widest part of the leaf . These three leaf grading traits can be used as the basis for identifying different apricot varieties, which will be of certain reference value for the future evaluation of leaf morphology of China's apricot germplasm resources and varietal identification.