Publication

Abstract : Image processing constitutes one of the most powerful and widely used computer science-related technologies, especially in the field of medical sciences. It is frequently employed to recognize and find early signs of many different cancers, including skin cancer. Then, the segmentation of skin cancer images remains a major concern for dermatologists due to inadequate lighting and other conditions associated with image acquisition. Therefore, this study addresses a new technique of diagnosing skin cancer via multilevel thresholding with enhanced fuzzy optimization. Firstly, the technique has taken two types of skin cancer images: benign and malignant skin cancer images. Precisely, in order to determine the optimal thresholds, fuzzy optimization is utilized on skin cancer images. Further, the cancerous images are thresholded using the obtained thresholds. Furthermore, using evaluation metrics, the performance of the proposed strategy and existing well-established techniques are compared with each other in the standard analysis. Finally, the experimental findings and comparisons reveal that the addressed technique improves consistency and segmentation quality when compared to current advanced techniques.