CONTRAST ENHANCEMENT WITH SALIENCY PRESERVATION
Author(s):
Vijaya Singh, Ankur Shukla, Sumit kr. Gupta
Keywords:
SSIM, PSNR, MSE, OBJECT
Abstract
For the evaluation of the quality of the enhanced image without human interventions, we need objective functions which tell us all about the quality of image through image parameters. Various objective functions are illustrated in this paper. To generate a good contrast enhancement image with preserved basic quantitative and qualitative parameters. Objective function formed here by combination of three performance measures, entropy value, contrast per pixel (CPP) and number of edges (edge pixels). Entropy gives the average information contained in image. If histogram is uniform that means information contain is high. and the contrast per pixel give the difference of the pixel brightness to the average brightness of the boundary pixel. If CPP is high then edge detection is easy. Entropy value gives the average information content of image or randomness present in image. If the distribution of the intensities (brightness level) in histogram of the image is uniform, then we can say that histogram is equalized and also the entropy of the image is high. The other objective functions considered are PSNR i.e., peak signal to noise ratio & MSE i.e., measure of enhancement, here SSIM i.e., structural similarity is low as in the case of enhancement the structural similarity is compromised for enhanced image. In images, edge contains vital information. Sharp edges are easily distinguishable and pleasing to human vision and also suitable for the machine learning. The object identification in the image needs sharp edges. The core idea the given literature is to maintain the mean brightness of the image, so image is illuminated properly. During enhancement the output image must have a high value of SSIM and also PSNR value must be high. So, the output image is not very different from input image. The brightness preservation in output image is measured using (AMBE) absolute mean brightness error value, image feature qualities i.e., similarity of output image to input image are checked by SSIM and PSNR. Contrast enhanced images from the technique achieves a good trade-off between focus in the image, low contrast improvement and brightness retaining along with the natural appeal of the input image.
Article Details
Unique Paper ID: 152947

Publication Volume & Issue: Volume 8, Issue 5

Page(s): 88 - 94
Article Preview & Download


Share This Article

Conference Alert

NCSST-2021

AICTE Sponsored National Conference on Smart Systems and Technologies

Last Date: 25th November 2021

SWEC- Management

LATEST INNOVATION’S AND FUTURE TRENDS IN MANAGEMENT

Last Date: 7th November 2021

Latest Publication

Go To Issue



Call For Paper

Volume 9 Issue 10

Last Date for paper submitting for March Issue is 25 March 2023

About Us

IJIRT.org enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on editor@ijirt.org

Social Media

Google Verified Reviews

Contact Details

Telephone:6351679790
Email: editor@ijirt.org
Website: ijirt.org

Policies

Important Links

Browse Rsearch papers