Objective For the existing subjective assessment methods for skin pruritus in patients with maintenance hemodialysis (MHD), which have problems of large individual differences and poor repeatability, an objective assessment and grading system based on image processing has been developed. Methods A total of 80 patients with MHD accompanied by skin pruritus, who were admitted to Xiaoshan District Hospital of Traditional Chinese Medicine in Hangzhou from January 2023 to December 2025, were selected as the research subjects. By means of image preprocessing, color space conversion, and feature extraction, an objective evaluation model was constructed. Moreover, the correlation between image feature parameters and subjective scores was analyzed, and a four-level pruritus grading model based on support vector machine (SVM) was established. Results The recognition accuracy of the image feature extraction model proposed in this paper was 96.8%, and the consistency of texture feature and scratch feature extraction were 94.2% and 92.5%, respectively. At the same time, the image feature parameters of the model are significantly positively correlated with visual analogue scale (VAS) score and four-items itching scale (FIIQ) total score. Apart from this, the SVM-based hierarchical model had an overall accuracy of 88.7% on the test set and a Kappa value of 0.82, which was in good agreement with clinician assessments. Conclusion The objective assessment and grading system based on image processing established in this study can effectively quantify the skin pruritus-related lesion characteristics of patients with MHD, providing a tool for clinical efficacy evaluation and individualized management.