Skin cancers image classification using transformation and first order statistic features with artificial neural network classifier

<span>Skin cancer is one of the most dangerous types cancer. Some this lead to death, so must be discovered and indexed avoid its spread through initial detection in impulsive stage. This paper deals with indexing different melanomas using an artificial neural network (ANN) depending on international skin imaging collaboration (ISIC) 2018 dataset that was used. The pre-processing important part because it formulates image by insolated from image. It consists four stages, removable, cropping, thinning, normalization. phase has been used eliminate all undesirable hair particles lesion. cropped transforms into frequency domain coefficients discrete cosine transform (DCT), wavelet (DWT), gradient for sub-band images extract feature. statistical feature extraction implemented minimize size data ANN training. experimental analysis ISIC consisting seven dermoscopic (this only). For classification purposes, accuracy obtained about 88.98% DWT, 85.44% DCT, 76.07% transform.</span>