New Proposed Fusion between DCT for Feature Extraction and NSVC for Face Classification

摘要

Feature extraction is an interactive and iterative analysis process of a largedataset of raw data in order to extract meaningful knowledge. In this article, wepresent a strong descriptor based on the Discrete Cosine Transform (DCT), we showthat the new DCT-based Neighboring Support Vector Classifier (DCT-NSVC)provides a better results compared to other algorithms for supervised classification.Experiments on our real dataset named BOSS, show that the accuracy ofclassification has reached 99%. The application of DCT-NSVC on MIT-CBCLdataset confirms the performance of the proposed approach.