Deep Learning Based Semantic Image Segmentation Methods for Classification of Web Page Imagery

Semantic segmentation is the task of clustering together parts an image that belong to same object class. webpages important for inferring contextual information from webpage. This study examines and compares deep learning methods classifying based on imagery obscured by semantic segmentation. Fully convolutional neural network architectures (UNet FCN-8) with defined hyperparameters loss functions are used demonstrate how they can support efficient method this type classification scenario in custom-prepared webpage data labeled multi-class semantically segmented masks using HTML elements such as paragraph text, images, logos, menus. Using proposed Seg-UNet model achieved best accuracy 95%. A comparison various optimizer demonstrates overall efficacy approach.