We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
UPRIGHT FAST-HARRIS FILTER.
- Authors
MOHAMMED, ABDULMALIK DANLAMI; MUHAMMED, ADAM; KOLO, IDRIS MOHAMMED; NDAKO, ADAMA VICTOR; ABDULHAMID, SHAFI'I MUHAMMED; HASSAN, ABDULKADIR BABA; MOHAMMED, ABUBAKAR SADDIQ
- Abstract
The traditional approaches to the classification of image regions suffer drawbacks in the face of imaging conditions (occlusion, illumination changes, rotation, viewpoint changes, and image blurring) and thus contribute to the poor performance of several vision based applications, such as object recognition, object tracking, image retrieval, pose estimation, camera calibration, 3D reconstruction, Structure from motion, stereo images, and image stitching. However, in this work, feature points extraction method by decomposition of image structure is employed in order to overcome these challenges. The decomposition of an image structure into feature set enhances the performance of many vision-based applications and system. The feature point extraction method which we refer to as Upright Feature from Accelerated Segment Test with Harris filter (UFAH) in this text, works by combining Feature from Accelerated Segment Test detector with Harris filter. The result obtained in the evaluation process shows that UFAH is robust and also invariant to imaging conditions (i.e rotation, illumination changes, and image blurring).
- Subjects
IMAGE processing; OBJECT recognition (Computer vision); PIXELS; IMAGING systems; MACHINE learning
- Publication
I-Manager's Journal on Image Processing, 2018, Vol 5, Issue 3, p14
- ISSN
2349-4530
- Publication type
Article
- DOI
10.26634/jip.5.3.15689