Contour-Based Corner Detection and Classification by Using Mean Projection Transform
Image corner detection is a fundamental task in computer vision. Many applications require reliable detectors to accurately detect corner points, commonly achieved by using image contour information. The curvature definition is sensitive to local variation and edge aliasing, and available smoothing...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2014-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/14/3/4126 |
| _version_ | 1817992343360372736 |
|---|---|
| author | Seyed Mostafa Mousavi Kahaki Md Jan Nordin Amir Hossein Ashtari |
| author_facet | Seyed Mostafa Mousavi Kahaki Md Jan Nordin Amir Hossein Ashtari |
| author_sort | Seyed Mostafa Mousavi Kahaki |
| collection | DOAJ |
| description | Image corner detection is a fundamental task in computer vision. Many applications require reliable detectors to accurately detect corner points, commonly achieved by using image contour information. The curvature definition is sensitive to local variation and edge aliasing, and available smoothing methods are not sufficient to address these problems properly. Hence, we propose Mean Projection Transform (MPT) as a corner classifier and parabolic fit approximation to form a robust detector. The first step is to extract corner candidates using MPT based on the integral properties of the local contours in both the horizontal and vertical directions. Then, an approximation of the parabolic fit is calculated to localize the candidate corner points. The proposed method presents fewer false-positive (FP) and false-negative (FN) points compared with recent standard corner detection techniques, especially in comparison with curvature scale space (CSS) methods. Moreover, a new evaluation metric, called accuracy of repeatability (AR), is introduced. AR combines repeatability and the localization error (Le) for finding the probability of correct detection in the target image. The output results exhibit better repeatability, localization, and AR for the detected points compared with the criteria in original and transformed images. |
| first_indexed | 2024-04-14T01:25:04Z |
| format | Article |
| id | doaj.art-0ae677bcf2dc44a98f871ca4f2a0c719 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-14T01:25:04Z |
| publishDate | 2014-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0ae677bcf2dc44a98f871ca4f2a0c7192022-12-22T02:20:29ZengMDPI AGSensors1424-82202014-02-011434126414310.3390/s140304126s140304126Contour-Based Corner Detection and Classification by Using Mean Projection TransformSeyed Mostafa Mousavi Kahaki0Md Jan Nordin1Amir Hossein Ashtari2Center for Artificial Intelligence Technology, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi, Selangor 43600, MalaysiaCenter for Artificial Intelligence Technology, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi, Selangor 43600, MalaysiaCenter for Artificial Intelligence Technology, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi, Selangor 43600, MalaysiaImage corner detection is a fundamental task in computer vision. Many applications require reliable detectors to accurately detect corner points, commonly achieved by using image contour information. The curvature definition is sensitive to local variation and edge aliasing, and available smoothing methods are not sufficient to address these problems properly. Hence, we propose Mean Projection Transform (MPT) as a corner classifier and parabolic fit approximation to form a robust detector. The first step is to extract corner candidates using MPT based on the integral properties of the local contours in both the horizontal and vertical directions. Then, an approximation of the parabolic fit is calculated to localize the candidate corner points. The proposed method presents fewer false-positive (FP) and false-negative (FN) points compared with recent standard corner detection techniques, especially in comparison with curvature scale space (CSS) methods. Moreover, a new evaluation metric, called accuracy of repeatability (AR), is introduced. AR combines repeatability and the localization error (Le) for finding the probability of correct detection in the target image. The output results exhibit better repeatability, localization, and AR for the detected points compared with the criteria in original and transformed images.http://www.mdpi.com/1424-8220/14/3/4126corner detectioncontour-based corner detectormean projection transformpolygonal approximation |
| spellingShingle | Seyed Mostafa Mousavi Kahaki Md Jan Nordin Amir Hossein Ashtari Contour-Based Corner Detection and Classification by Using Mean Projection Transform Sensors corner detection contour-based corner detector mean projection transform polygonal approximation |
| title | Contour-Based Corner Detection and Classification by Using Mean Projection Transform |
| title_full | Contour-Based Corner Detection and Classification by Using Mean Projection Transform |
| title_fullStr | Contour-Based Corner Detection and Classification by Using Mean Projection Transform |
| title_full_unstemmed | Contour-Based Corner Detection and Classification by Using Mean Projection Transform |
| title_short | Contour-Based Corner Detection and Classification by Using Mean Projection Transform |
| title_sort | contour based corner detection and classification by using mean projection transform |
| topic | corner detection contour-based corner detector mean projection transform polygonal approximation |
| url | http://www.mdpi.com/1424-8220/14/3/4126 |
| work_keys_str_mv | AT seyedmostafamousavikahaki contourbasedcornerdetectionandclassificationbyusingmeanprojectiontransform AT mdjannordin contourbasedcornerdetectionandclassificationbyusingmeanprojectiontransform AT amirhosseinashtari contourbasedcornerdetectionandclassificationbyusingmeanprojectiontransform |