IMAGE BACKGROUND MATCHING FOR IDENTIFYING SUSPECTS

摘要

Thousands of digital images may exist of a given location, some of which may show a crime in progress. One technique for identifying suspects and witnesses is to collect images of specific crime scenes from computers, cell phones, cameras and other electronic devices, and perform image matching based on image backgrounds. This paper describes an image matching technique that is used in conjunction with feature generation methodologies, such as the Scale Invariant Feature Transform (SIFT) and the Speeded Up Robust Features (SURF) algorithms. The technique identifies keypoints in images of a given location with minor differences in viewpoint and content. After calculating keypoints for the images, the technique stores only the "good" features for each image to minimize space and matching requirements. Test results indicate that matching accuracy exceeding 80% is obtained with the SIFT and SURF algorithms.