The trace transform for object identification
We know that Trace transform is is a novel algorithm with properties like scale and rotation invariance and occlusion robustness [1] that has been shown to be effective in recognition tasks using generalisation of Radon transform. However its computation complexity has been a problem. A Trace transf...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/48510 |
| _version_ | 1824456683784503296 |
|---|---|
| author | Leong, Wei Siong. |
| author2 | School of Computer Engineering |
| author_facet | School of Computer Engineering Leong, Wei Siong. |
| author_sort | Leong, Wei Siong. |
| collection | NTU |
| description | We know that Trace transform is is a novel algorithm with properties like scale and rotation invariance and occlusion robustness [1] that has been shown to be effective in recognition tasks using generalisation of Radon transform. However its computation complexity has been a problem. A Trace transform of the image is a 2D function of the parameters of each tracing line.
In this project, exploring the uses of Hough transform which extract primitives, such as lines, edges and curves, is often a key step in an image analysis procedure. We designed a function that take in Polar Coordinates (r,θ) inputs and converting to Cartesian Coordinates (x,y). As Hough transform is the simplest example for detection of lines.
Experiment result from this function will allow us to get all the pixel and its coordinate from a image base on the given inputs. |
| first_indexed | 2025-02-19T03:58:00Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/48510 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:58:00Z |
| publishDate | 2012 |
| record_format | dspace |
| spelling | ntu-10356/485102023-03-03T20:28:23Z The trace transform for object identification Leong, Wei Siong. School of Computer Engineering Suhaib A Fahmy DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision We know that Trace transform is is a novel algorithm with properties like scale and rotation invariance and occlusion robustness [1] that has been shown to be effective in recognition tasks using generalisation of Radon transform. However its computation complexity has been a problem. A Trace transform of the image is a 2D function of the parameters of each tracing line. In this project, exploring the uses of Hough transform which extract primitives, such as lines, edges and curves, is often a key step in an image analysis procedure. We designed a function that take in Polar Coordinates (r,θ) inputs and converting to Cartesian Coordinates (x,y). As Hough transform is the simplest example for detection of lines. Experiment result from this function will allow us to get all the pixel and its coordinate from a image base on the given inputs. Bachelor of Engineering (Computer Science) 2012-04-25T05:51:49Z 2012-04-25T05:51:49Z 2012 2012 Final Year Project (FYP) http://hdl.handle.net/10356/48510 en Nanyang Technological University 20 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Leong, Wei Siong. The trace transform for object identification |
| title | The trace transform for object identification |
| title_full | The trace transform for object identification |
| title_fullStr | The trace transform for object identification |
| title_full_unstemmed | The trace transform for object identification |
| title_short | The trace transform for object identification |
| title_sort | trace transform for object identification |
| topic | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision |
| url | http://hdl.handle.net/10356/48510 |
| work_keys_str_mv | AT leongweisiong thetracetransformforobjectidentification AT leongweisiong tracetransformforobjectidentification |