Computer Vision based Skin Cancer Detection
In this project I have used the well-known data set which is used by most of the scholars who are working on the skin cancer related projects. The main reason of me and other researchers for selecting this dataset is that as the quality of images is very good, properly cropped and they are very well labelled. The short size and dearth of diversity of the collection of dermatoscopic images available hinder the training of neural networks for automated diagnosis of pigmented skin lesions. The HAM10000 ("Human Against Machine with 10,000 training images") dataset is how we address this issue. We gathered dermatoscopic images from various populations that were captured and saved using various modalities. 10015 dermatoscopic images make up the final dataset, which can be used as a training set for research-based machine learning. Actinic keratoses andintraepithelial carcinoma / Bowen’s disease (akiec), basal cell carcinoma (bcc), benign keratosis-like lesions (solar lentigines / seborrheic keratoses and lichen-planus like keratoses, bkl), dermatofibroma (df), melanoma (mel), melanocytic nevi (nv), and vascular lesions (angiomas, angiokeratomas, pyogenic granulomas and hemorrhage, vasc) are also represented in the cases [17].
Due to the restriction of uploading the data, they have stored them into two file which are as follows: HAM10000_images_part1.zip(5000 images files in JPEG form) HAM10000_images_part2.zip(5015 images files in JPEG form)