Skin cancer is a deadly disease, which require early detection to prevent and treat it successfully in its early stage. This paper address the detection of skin cancer using Deep Convolutional Neural Network (DCNNs) classification approach.The classification stage follows by implementing wiener filter to remove the impacts of noise and inverts the blurring in the input image .Further, Segmentation is done using k-means clustering which separates the interest area from the background to isolate the region of interest. The proposed method uses, Scale Invariant Feature Transform (SIFT) for feature extraction to identify unique and reliable characteristics in an image to increase the deep learning model's effectiveness and precision. For the accurate classification of skin cancer, a DCNNs model is analysed where the architecture combine Convolutional and Recurrent Neural Network (CNN-RNN) to enhance skin cancer classification. The proposed method is developed in python software with high accuracy of 93% makes the proposed methodology to be efficient and wellbeing.