Mallik, Sudhansu (2016) Underwater Image Enhancement. MTech thesis.
|PDF (Full text is restricted upto 26.04.2020) |
Restricted to Repository staff only
The human incapability in diving in the deep ocean for a long time has increased the challenges of underwater analysis. Underwater remotely operated vehicles(UROVs) are used these days to acquire images in deep ocean. However, the challenges contain problems like disrupting scattering effect and aggressive light absorption in the deep ocean. Although UROVs are having artificial light, it is unable to provide good and clear images because of these problems. Three methods have been discussed in the thesis to improve the image quality. first of all, a method which contains a haze removal algorithm followed by a Contrast Limited Adaptive Histogram Equalization (CLAHE) color model has been discussed. The underwater images are enhanced through haze removal algorithm by dark channel prior technique. It shows a good result by reducing haze and noise effect still, it has a tendency to darken the image in some situation. CLAHE on RGB model has been followed in our approach to change the level of contrast and intensity of dehaze image. Secondly, an algorithm has been proposed based on Empirical Mode Decomposition (EMD) with a white balanced input to enhance the visual quality of the under-water images. The image is processed through the Gray World technique which is a white balance approach to enhance the contrast of the image and to remove the unwanted color cast in the image. Each R,G and B channel of the resultant image is decomposed into its Intrinsic Mode Functions (IMFs) by EMD process. Final enhanced image has been constructed by combining the IMFs of each channel with different optimised weights. In the end, a novel image enhancement algorithm is described for underwater images using fusion process. To improve the quality of the degraded image and to overcome from the limitations of underwater, two inputs that represent white balanced and color corrected versions of the original image have been defined. Gray World is a white balance algorithm characterized to overcome from the unwanted color casts has been implemented. To enhance the visibility of the the degraded portion, four weight maps are considered. Then each map of the resultant image has decomposed into its Intrinsic Mode Functions(IMFs) using EMD. Our experimental results show that the proposed algorithm has significantly improved quality of underwater images by enhancing the contrast of the image and reducing noise as well as artifacts in the image.
|Item Type:||Thesis (MTech)|
|Uncontrolled Keywords:||DCP; CLAHE; Haze Removal; Gray-World; IMFs; EMD; Weights; Color Correction; Image Enhancements|
|Subjects:||Engineering and Technology > Electronics and Communication Engineering > Image Processing|
|Divisions:||Engineering and Technology > Department of Electronics and Communication Engineering|
|Deposited By:||Mr. Sanat Kumar Behera|
|Deposited On:||27 Apr 2018 20:40|
|Last Modified:||27 Apr 2018 20:40|
|Supervisor(s):||Pati, Umesh Chandra|
Repository Staff Only: item control page