Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination
The study investigates automatic inspection and processing of the stem accessories based on vision stitching and spectral illumination. The vision stitching mainly involves algorithms of white balance, scale-invariant feature transforms (SIFT) and roundness for whole image of automatic accessory ins...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Chinese Institute of Automation Engineers (CIAE) & Taiwan Smart Living Space Association (SMART LISA)
2014-08-01
|
| Series: | International Journal of Automation and Smart Technology |
| Subjects: | |
| Online Access: | http://www.ausmt.org/index.php/AUSMT/article/view/496 |
| _version_ | 1811331695845048320 |
|---|---|
| author | Wen-Yang Chang |
| author_facet | Wen-Yang Chang |
| author_sort | Wen-Yang Chang |
| collection | DOAJ |
| description | The study investigates automatic inspection and processing of the stem accessories based on vision stitching and spectral illumination. The vision stitching mainly involves algorithms of white balance, scale-invariant feature transforms (SIFT) and roundness for whole image of automatic accessory inspection. The illumination intensities, angles, and spectral analyses of light sources are analyzed for image optimal inspections. The unrealistic color casts of feature inspection is removed using a white balance algorithm for global automatic adjustment. The SIFT is used to extract and detect the image features for big image stitching. The Hough transform is used to detect the parameters of a circle for roundness of the bicycle accessories. The feature inspections of a stem contain geometry size, roundness, and image stitching. Results showed that maximum errors of 0°, 10°, 30°, and 50° degree for the spectral illumination of white light LED arrays with differential shift displacements are 4.4, 4.2, 6.8, and 3.5 %, respectively. The deviation error of image stitching for the stem accessory in x and y coordinates are 2 pixels. The SIFT and RANSAC enable to transform the stem image into local feature coordinates. |
| first_indexed | 2024-04-13T16:24:32Z |
| format | Article |
| id | doaj.art-30e6ad3f1b8145c2ae4949a03f3eb436 |
| institution | Directory Open Access Journal |
| issn | 2223-9766 |
| language | English |
| last_indexed | 2024-04-13T16:24:32Z |
| publishDate | 2014-08-01 |
| publisher | Chinese Institute of Automation Engineers (CIAE) & Taiwan Smart Living Space Association (SMART LISA) |
| record_format | Article |
| series | International Journal of Automation and Smart Technology |
| spelling | doaj.art-30e6ad3f1b8145c2ae4949a03f3eb4362022-12-22T02:39:47ZengChinese Institute of Automation Engineers (CIAE) & Taiwan Smart Living Space Association (SMART LISA)International Journal of Automation and Smart Technology2223-97662014-08-014314314910.5875/ausmt.v4i3.496130Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral IlluminationWen-Yang Chang0National Formosa UniversityThe study investigates automatic inspection and processing of the stem accessories based on vision stitching and spectral illumination. The vision stitching mainly involves algorithms of white balance, scale-invariant feature transforms (SIFT) and roundness for whole image of automatic accessory inspection. The illumination intensities, angles, and spectral analyses of light sources are analyzed for image optimal inspections. The unrealistic color casts of feature inspection is removed using a white balance algorithm for global automatic adjustment. The SIFT is used to extract and detect the image features for big image stitching. The Hough transform is used to detect the parameters of a circle for roundness of the bicycle accessories. The feature inspections of a stem contain geometry size, roundness, and image stitching. Results showed that maximum errors of 0°, 10°, 30°, and 50° degree for the spectral illumination of white light LED arrays with differential shift displacements are 4.4, 4.2, 6.8, and 3.5 %, respectively. The deviation error of image stitching for the stem accessory in x and y coordinates are 2 pixels. The SIFT and RANSAC enable to transform the stem image into local feature coordinates.http://www.ausmt.org/index.php/AUSMT/article/view/496IlluminationAutomatic InspectionLight SourceVision Stitching |
| spellingShingle | Wen-Yang Chang Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination International Journal of Automation and Smart Technology Illumination Automatic Inspection Light Source Vision Stitching |
| title | Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination |
| title_full | Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination |
| title_fullStr | Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination |
| title_full_unstemmed | Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination |
| title_short | Automatic Inspection and Processing of Accessory Based on Vision Stitching and Spectral Illumination |
| title_sort | automatic inspection and processing of accessory based on vision stitching and spectral illumination |
| topic | Illumination Automatic Inspection Light Source Vision Stitching |
| url | http://www.ausmt.org/index.php/AUSMT/article/view/496 |
| work_keys_str_mv | AT wenyangchang automaticinspectionandprocessingofaccessorybasedonvisionstitchingandspectralillumination |