Publication: Optimisation of object detection and measurement using matlab stereo vision
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering | |
| dc.contributor.author | Aswan, Muhamad Afnan Naqiuddin | |
| dc.date.accessioned | 2024-02-27T06:56:10Z | |
| dc.date.available | 2024-02-27T06:56:10Z | |
| dc.date.issued | 2022-08-01 | |
| dc.description.abstract | In the object detection and measurement, camera sensors can be very powerful and if compared to the infrared or ultrasonic sensors. With the aid of computer vision, the detection may provide additional knowledge about the environment, such as the distance between an object and its features such as its dimension and measurement. Here, stereo vision is being selected in this project as it can perform both object detection and measurement. The calibration of the stereo system was done to obtain the parameters required for the image rectification in the disparity mapping process. In this project, three stereo matching algorithms were examined to find the best disparity map. Here, the Normal Matching algorithm provide the best result of the disparity map representation and was chosen in this project compared to the Semi-Global Matching and Block Matching algorithm. The disparity value from the mapping process was extracted. In the object detection, the blob detection in the binary image was performed and the square box on the detected object was drawn. The pixel value was extracted and the length per pixel constant was calculated. Then, optimised equations for the depth width, height measurement was generated, and the performance of the optimised object measurement was analysed. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/18477 | |
| dc.language.iso | en | |
| dc.title | Optimisation of object detection and measurement using matlab stereo vision | |
| dc.type | Resource Types::text::report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |