In this study, 104 sugar beet germplasm resources from Northeast China were evaluated with the aim of screening elite sugar beet germplasm and providing a theoretical basis for further germplasm improvement and breeding. Principal component analysis, correlation analysis, genetic diversity analysis and cluster analysis were performed on 22 traits. The results showed that the coefficients of variation for the 13 qualitative traits ranged from 0.96% to 99.04%, with genetic diversity indices ranging from 0.05 to 1.07, while the coefficients of variation for the 9 quantitative traits ranged from 8.02% to 55.74%, with genetic diversity indices ranging from 1.88 to 2.27. Among these traits, cercospora leaf spot exhibited the largest variation index, whereas growth vigor had the smallest coefficient of variation. The 22 traits were reduced to six principal components by principal component analysis, and the 104 sugar beet germplasm resources were classified into four groups by cluster analysis. There were large genetic differences among the sugar beet germplasm resources. Growth vigor, plant height, petiole width, petiole length, root furrow depth, root shape, root skin color and leaf color were identified as important factors affecting phenotypic variation and yield in sugar beet and can be used as major morphological indicators for parent selection and germplasm evaluation.