New shape descriptor in the context of edge continuity
The object contour is a significant cue for identifying and categorising objects. The current work is motivated by indicative researches that attribute object contours to edge information. The spatial continuity exhibited by the edge pixels belonging to the object contour make these different from t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2019-04-01
|
| Series: | CAAI Transactions on Intelligence Technology |
| Subjects: | |
| Online Access: | https://digital-library.theiet.org/content/journals/10.1049/trit.2019.0002 |
| _version_ | 1819182808082415616 |
|---|---|
| author | Seba Susan Prachi Agrawal Prachi Agrawal Minni Mittal Srishti Bansal |
| author_facet | Seba Susan Prachi Agrawal Prachi Agrawal Minni Mittal Srishti Bansal |
| author_sort | Seba Susan |
| collection | DOAJ |
| description | The object contour is a significant cue for identifying and categorising objects. The current work is motivated by indicative researches that attribute object contours to edge information. The spatial continuity exhibited by the edge pixels belonging to the object contour make these different from the noisy edge pixels belonging to the background clutter. In this study, the authors seek to quantify the object contour from a relative count of the adjacent edge pixels that are oriented in the four possible directions, and measure using exponential functions the continuity of each edge over the next adjacent pixel in that direction. The resulting computationally simple, low-dimensional feature set, called as ‘edge continuity features’, can successfully distinguish between object contours and at the same time discriminate intra-class contour variations, as proved by the high accuracies of object recognition achieved on a challenging subset of the Caltech-256 dataset. Grey-to-RGB template matching with City-block distance is implemented that makes the object recognition pipeline independent of the actual colour of the object, but at the same time incorporates colour edge information for discrimination. Comparison with the state-of-the-art validates the efficiency of the proposed approach. |
| first_indexed | 2024-12-22T22:52:00Z |
| format | Article |
| id | doaj.art-b4bc52e1c17e4bbabf3bbeca0e2cf798 |
| institution | Directory Open Access Journal |
| issn | 2468-2322 |
| language | English |
| last_indexed | 2024-12-22T22:52:00Z |
| publishDate | 2019-04-01 |
| publisher | Wiley |
| record_format | Article |
| series | CAAI Transactions on Intelligence Technology |
| spelling | doaj.art-b4bc52e1c17e4bbabf3bbeca0e2cf7982022-12-21T18:09:56ZengWileyCAAI Transactions on Intelligence Technology2468-23222019-04-0110.1049/trit.2019.0002TRIT.2019.0002New shape descriptor in the context of edge continuitySeba Susan0Prachi Agrawal1Prachi Agrawal2Minni Mittal3Srishti Bansal4Department of Information Technology, Delhi Technological UniversityDepartment of Information Technology, Delhi Technological UniversityDepartment of Information Technology, Delhi Technological UniversityDepartment of Information Technology, Delhi Technological UniversityDepartment of Information Technology, Delhi Technological UniversityThe object contour is a significant cue for identifying and categorising objects. The current work is motivated by indicative researches that attribute object contours to edge information. The spatial continuity exhibited by the edge pixels belonging to the object contour make these different from the noisy edge pixels belonging to the background clutter. In this study, the authors seek to quantify the object contour from a relative count of the adjacent edge pixels that are oriented in the four possible directions, and measure using exponential functions the continuity of each edge over the next adjacent pixel in that direction. The resulting computationally simple, low-dimensional feature set, called as ‘edge continuity features’, can successfully distinguish between object contours and at the same time discriminate intra-class contour variations, as proved by the high accuracies of object recognition achieved on a challenging subset of the Caltech-256 dataset. Grey-to-RGB template matching with City-block distance is implemented that makes the object recognition pipeline independent of the actual colour of the object, but at the same time incorporates colour edge information for discrimination. Comparison with the state-of-the-art validates the efficiency of the proposed approach.https://digital-library.theiet.org/content/journals/10.1049/trit.2019.0002image representationfeature extractionobject recognitionedge detectionimage colour analysislearning (artificial intelligence)edge continuity featuresobject contourintra-class contour variationsobject recognition pipelinecolour edge informationidentifying categorising objectsnoisy edge pixelsadjacent edge pixelsadjacent pixel |
| spellingShingle | Seba Susan Prachi Agrawal Prachi Agrawal Minni Mittal Srishti Bansal New shape descriptor in the context of edge continuity CAAI Transactions on Intelligence Technology image representation feature extraction object recognition edge detection image colour analysis learning (artificial intelligence) edge continuity features object contour intra-class contour variations object recognition pipeline colour edge information identifying categorising objects noisy edge pixels adjacent edge pixels adjacent pixel |
| title | New shape descriptor in the context of edge continuity |
| title_full | New shape descriptor in the context of edge continuity |
| title_fullStr | New shape descriptor in the context of edge continuity |
| title_full_unstemmed | New shape descriptor in the context of edge continuity |
| title_short | New shape descriptor in the context of edge continuity |
| title_sort | new shape descriptor in the context of edge continuity |
| topic | image representation feature extraction object recognition edge detection image colour analysis learning (artificial intelligence) edge continuity features object contour intra-class contour variations object recognition pipeline colour edge information identifying categorising objects noisy edge pixels adjacent edge pixels adjacent pixel |
| url | https://digital-library.theiet.org/content/journals/10.1049/trit.2019.0002 |
| work_keys_str_mv | AT sebasusan newshapedescriptorinthecontextofedgecontinuity AT prachiagrawal newshapedescriptorinthecontextofedgecontinuity AT prachiagrawal newshapedescriptorinthecontextofedgecontinuity AT minnimittal newshapedescriptorinthecontextofedgecontinuity AT srishtibansal newshapedescriptorinthecontextofedgecontinuity |