Spatial Semantic Images with Relationship Contents by Using Convolutional Neural Network and Support Vector Machine

Abstract

In recently, semantic image is an active problem in the digital image processing field. A large number of new techniques and systems have researcher involved and attempted to improve the problems. The most of techniques is done by keyword searching model. Therefore, we propose a new approach to classify the relationships between object and action. The approach is composed of three main phases: (1) data preprocessing, (2) relationship between contents, and (3) measurement and evaluation. We train and test our model on a largescale image dataset of actions. The major information contents use the relationships between object and action. The results indicated that the proposed method offers significant performance improvements in semantic classification with a maximum success rate of 80.9%.