Skin Cancer Detection Using Image Processing and Deep Learning

Skin cancer has become one of the most common and dangerous diseases affecting humans worldwide. Early detection plays a vital role in improving treatment outcomes and survival rates. This research proposes a novel skin cancer detection system that utilizes image processing and deep learning methods to identify cancerous lesions efficiently. The system processes dermatological images through Fourier spectral analysis and texture- based segmentation, extracting key features using the Gray Level Co-occurrence Matrix (GLCM) technique. These features are then classified using a Convolutional Neural Network (CNN) model trained on a dataset of 10,000 skin images sourced from Kaggle. The CNN distinguishes seven major skin lesion categories including melanoma, basal cell carcinoma, actinic keratoses, and benign keratosis. The system achieved an accuracy level of 95.4%, demonstrating the potential of machine learning in early and reliable skin cancer diagnosis. This automated approach reduces human error, saves time, and enhances diagnostic precision, making it a promising tool for clinical dermatology applications.