Skin recognition is the process of finding flesh-coloured pixels and regions in an image. It is generally used as a preliminary step to look for specific areas in the image that could possibly have human faces and limbs. There have been numerous computer vision approaches that have been developed for skin recognition. Skin segmentation assumes an important role in several applications like medical X Ray imaging (MXRI) and magnetic resonance imaging(MRI). Other applications of skin segmentation include surveillance, human computer interaction, face recognition, etc. There is no certainty that the regions detected correspond to only skin. Hence, to say that the colour of the detected region and that of the skin is the same, is a definitive. For detecting the face in the image, the regions which are not the same colour as skin can be eliminated reliably as they are not necessary for our process. A thresholding process should be used to segment the skin regions from the rest of the image as usually the skin regions would be brighter than the other regions. Among all the supervised learning algorithms, back propagation (BP) is probably the most widely used. We have used back propagation for carrying the skin segmentation of the input image. The dataset source is taken from the UCI Repository. The dataset is collected by randomly sampling R, G, B values from the images. These RGB values are then mapped to the HSV colour model for a more adaptive and effective detection. The success rate of the implementation is 94.317568%. © IAEME Publication.