Image Enhancement using Modified Laplacian Filter, Clahe and Adaptive Thresholding

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, a new method for image enhancement is proposed for object recognition in autonomous vehicles. The new method involves image segmentation, Contrast-Limited Advanced Histogram Equalization (CLAHE), a Modified Laplacian filter and Adaptive Thresholding. The image is segmented into four-by-four segments. Each segment is assessed and corrected for blurriness and darkness. Adaptive Thresholding gives a unique automated threshold value for each region of the image depending upon the lighting conditions of the region of the image. The aim is to enhance the contrast and brightness of the image and increase sharpness of the image without inducing noise. The proposed method enhances the image quality that the image recognition system receives and was tested on low-light datasets. The extensive experiments show higher performance both qualitative and quantitively compared to the state of art methods.
Original languageEnglish
Title of host publication2024 International Conference on Intelligent Systems and Computer Vision (ISCV)
Place of PublicationFez, Morocco
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Number of pages7
ISBN (Electronic)979-8-3503-5018-0
DOIs
Publication statusPublished - 12 Aug 2024
Event2024 International Conference on Intelligent Systems and Computer Vision (ISCV) - Fez, Morocco
Duration: 8 May 202410 May 2024
https://www.iscvconf.com/2024/

Publication series

Name2024 International Conference on Intelligent Systems and Computer Vision, ISCV 2024

Conference

Conference2024 International Conference on Intelligent Systems and Computer Vision (ISCV)
Abbreviated titleIEEE ISCV 2024
Country/TerritoryMorocco
CityFez
Period8/05/2410/05/24
Internet address

Keywords

  • Image segmentation
  • CLAHE
  • Laplacian filter
  • adaptive thresholding
  • image recognition
  • and image recognition

Cite this