Early and accurate detection of skin diseases is essential for effective treatment. However, there is often a gap between dermatologists and patients due to a lack of knowledge about symptoms and stages of skin conditions. This study introduces an innovative dual deep learning-based classifierfor skin disease classification model that aims to bridge this gap using machine learning techniques. The model undergoes various preprocessing steps, such as image resizing, hair removal, and noise reduction, to improve image quality. It then utilizes the Mask Residual Convolutional Neural Network (Mask-RCNN) architecture for instance segmentation to isolate lesion regions in skin images. Color and texture features are extracted from the segmented lesions and input into the proposed Dual Deep Belief Network (DDBN) classifiers for concurrently classifying multiple skin diseases. Experiments conducted on the ISIC 2019 Challenge and HAM10000 datasets demonstrate the DDBN's effectiveness, achieving high accuracy, precision, recall, and F1-score compared to the existingclassification models. The DDBN's robust classification capabilities can aid dermatologists and patients in the early detection of skin diseases for timely intervention.