000			01968nam a22002417a 4500
003			ft8913
005			20251218154657.0
008			251218b ||||| |||| 00| 0 eng d
041			_aengtag
050			_aQA76.9 A43 A43 2025
082			_a.
100	1		_aAmadeo III, Alfedo M.; Bautista, Bella Marie B.; Lingad, Alexandei Paul A.
245			_aAn enhancement of Estacio-Laurente oriented fast and rotated brief (ORB) algorithm for non-real-time image-based product research
264		1	_a. _b. _cc2025
300			_bUndergraduate Thesis: (Bachelor of Science in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2025
336			_2text _atext _btext
337			_2unmediated _aunmediated _bunmediated
338			_2volume _avolume _bvolume
505			_aABSTRACT: The Estacio-Laurente ORB algorithm is an improved version of the traditional ORB however, it can still detect key points from image noises and it cannot emphasize image edges that can help in detecting distinct features in that area. It can also retain only up to 1000 key points for feature matching. This limits the ability of the ORB to match all the possible distinct points an image can have. The last problem was the mismatches produced by the FLANN matcher. The existing algorithm does not have the ability to filter out outliers. The enhanced method is to improve the quality of the detected features, use the maximum possible number of detected features for matching, and improve the feature matching accuracy of the algorithm. To do this, methods such as Gaussian Blur, Canny Edge, Bayesian Optimization, and MAGSAC++ are used. The result shows that the changes in the existing algorithm improved the ability to detect good quality of features, it can detect and retain more distinct key points for feature matching, and it improves the accuracy and reliability of the matches as it is able to filter out outliers.
526			_aF
655			_aacademic writing
942			_2lcc _cMS
999			_c37380 _d37380