Bigdata Testing Using ETL and Digital Image Processing

Abstract: Melanoma is considered to be a fatal form of skin cancer. However, due to their similar visual look and symptoms, it is often difficult to differentiate it from the nevus. The number of cases among children is growing but if it is detected at its earlier stage then the risk of survival is very high. The cost and time needed for the physicians to treat all melanoma patients is very high. In this research work we propose a method that uses state-of-the-art image processing techniques to detect and differentiate melanoma from nevus. Pre processing is initially used to eliminate noise from the skin lesion of the images obtained, followed by the use of enhanced HSV colour space conversion clump to section out the lesion. The extraction of textural and colour choices from the lesion Shape a distinctive hybrid super-feature vector. Support Vector Machine (SVM) is used to classify skin cancer into melanoma and nevus. Our goal is to check the efficacy of the projected segmentation strategy, select the most suitable options and compare the results of the classification Within the literature the opposite strategies are present. Our proposed methodology archives encouraging result

Keywords: Support Vector Machine (SVM), Image processing techniques, distinctive hybrid super-feature vector.