Underwater Video Enhancement Using Enhanced Integra Ted Color Model Algorithm And Multithreading Technique
| dc.contributor.author | Neima, Hikmat z | |
| dc.date.accessioned | 2016-10-25T02:43:31Z | |
| dc.date.available | 2016-10-25T02:43:31Z | |
| dc.date.issued | 2009-06 | |
| dc.description.abstract | Aquatic studies that involve the classification of underwater organisms such as plants and animals are usually highly dependent on images and videos captured. In underwater images and videos, the blue color is often the dominant color. Nonuniform lighting, low contrast and diminished color are examples of underwater environmental effects. Underwater images and videos are of poor quality due to the nature of water and its density which is denser than air by approximately 800 times. The main aim of this research isoto enhance the existing Integrated Color Model (ICM) algorithm which is mainly used for underwater image enhancement. The enhancement was done by making the stretching parameters adaptive rather than using constant parameters. This was done by testing the current image to determine which color was dominant and the stretching parameter will depend on this. The enhanced ICM was then compared to the existing ICM using number of edges generated by both algorithms and also by their histograms. For 50 images, the results shown that the ICM generated images with 286731 edges while the Enhanced ICM generated 306416 edges. Secondly, the research introduced the use of a shared memory programming model to minimize the time required to enhance underwater images and videos by the enhanced ICM. The results obtained showed that when the image size (frame size in terms of video) was larger than 1 MByte and the number of threads was three, the benefit of parallelization was more noticeable. | en_US |
| dc.identifier.uri | http://hdl.handle.net/123456789/2872 | |
| dc.subject | Underwater images and videos are of poor quality | en_US |
| dc.subject | due to the nature of water | en_US |
| dc.title | Underwater Video Enhancement Using Enhanced Integra Ted Color Model Algorithm And Multithreading Technique | en_US |
| dc.type | Thesis | en_US |
Files
License bundle
1 - 1 of 1
Loading...
- Name:
- license.txt
- Size:
- 1.71 KB
- Format:
- Item-specific license agreed upon to submission
- Description: