A Novel Algorithm for Contrast Enhancement for Color Videos

Shaik Sharmila, Mary Junitha

Abstract


Existing methods are not giving satisfactory results for contrast enhancement as there is lot of parameters which should get match with required implementation algorithm. This implementation is totally based on the famous image processing tool contrast enhancement technique for enhancing the quality of colour videos captured under different deformations such as poor illumination and varying environmental conditions,etc. As we know the Video is nothing but frames of images, we will convert video into frames of images. The sequences of Images are converted from RGB color space to HSV colour space where we can get enhancement and get back to the RGB color space. The basic Class Limited Adaptive Histogram Equalization (CLAHE) is also used for enhancing the luminance component (V). Discrete Wavelet Transform(DWT) which  is very famous transform techniques applied to the Saturation (S) components, after that the decomposed approximation coefficients are changed with the help of mapping function which is got with the help of scaling triangle transform. The enhancement in the S component can get by with the help of Inverse Wavelet transforms(IDWT). The image obtained after this step is then converted back to the original RGB images. Subjective (visual quality inspection) as well as objective parameters  such as Peak-signal-to-noise ratio (PSNR), Absolute Mean Brightness Error (AMBE) and Mean squared error (MSE)) were used to check the performance of the system.The algorithm implemented can be compared with existing state of art  techniques  which will show the robustness of proposed work.The proposed work will give the better results with comparision. The same implementation is applied for video and we will get the improved contrast enhancement for real time videos. Video is nothing but frames of images.


Full Text:

PDF




Copyright (c) 2018 Edupedia Publications Pvt Ltd

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

 

All published Articles are Open Access at  https://journals.pen2print.org/index.php/ijr/ 


Paper submission: ijr@pen2print.org