The accurate assessment of skin barrier impairment is of significant importance in guiding skin-related care and treatment. TEWL(trans epidermal water loss) is considered the gold standard for evaluating skin barrier damage, but its measurement is complex and costly. Dermoscopy images offer advantages such as simplicity, low cost, non-invasiveness, and real-time capabilities. The features of dermoscopy images related to impaired skin barrier were investigated. Dermoscopy images were initially preprocessed using white balance, grayscaling, histogram equalization, and Wiener filtering. Subsequently, statistical features, texture features, and edge features of the images were extracted and analyzed for correlation with TEWL detection results in a skin-damaged mouse model. The results indicate a high correlation between texture features and TEWL values, especially the entropy of local binary patterns with a radius of 1, which has the highest correlation coefficient of 0.752. The texture features of dermoscopy images are found to effectively depict the degree of skin barrier damage, suggesting their potential in developing new methods for skin barrier function assessment as an alternative to TEWL.