Recognition of tongue print biometrie using binary robust independent elementary features

The study presents a tongue print biometrie recognition system that can use both SIFT keypoint descriptor and BRIEF keypoint descriptor algorithms. The main purpose of the study is to compare which of the two algorithms has faster recognition speed. The system captures tongue print images using a Raspberry Pi Camera. After image capture, the image is pre-processed using Contrast Limited Adaptive Histogram Equalization. SIFT feature extractor is then applied to the image to extract its features. The descriptor computation used both SIFT and BRIEF and they descriptors computed are stored in a database together with a unique user ID. The unique user ID used in the C# program to search the database for the user's information. Sample size of (30) thirty user was used for testing the proposed system. The test results show that using the BRIEF algorithm for tongue print recognition has an average recognition speed of 7.644 seconds while the SIFT algorithm's 13.829 seconds. The accuracy test results show that using the BRIEF algorithm also results to an improvement of recall, precision and accuracy over the SIFT algorithm.