VEHICLE CLASSIFICATION FROM CCTV IMAGES AND VIDEOS

• Main Author: KAMARUDDIN, ERNA SHAFFIQA
• Format: Final Year Project
• Language: English
• Published: 2018
• Online Access: http://utpedia.utp.edu.my/19164/1/FINAL%20REPORT_20782.pdf; http://utpedia.utp.edu.my/19164/

This report shows a study on orientation robust vehicle classification for data analytics purposes by using (HOG) feature set. The study is conducted to analyse the performance of histograms of oriented gradient(HOG) between Linear SVM algorithm and Aggregated Channel Features (ACF) algorithm which is using Decision Trees. The method and experimental design is shown to demonstrate the good accuracy for detecting and classifying object with various angle. The feature extraction obtain is trained using Linear SVM and Decision Trees. The result shows that the Linear SVM algorithm outperform ACF algorithm with accuracy percentage of 88.5% compared to 62.8% for smaller datasets.