PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 3, 2026Discover ComputingOpen Access

Sky-feature-constrained atmospheric light estimation with real-time ZYNQ implementation

View Full Paper
Ask AI
Bookmark
Share

Authors

XZX W ZhaoQWQiang WuJCJ Y Chen

Discussion

Loading...

Member takes

Overview

Randomized trial finds improved dehazing in outdoor images using sky-feature constraints, indicating enhanced visibility.

Key Points

  • The aim is to improve visibility in outdoor images by developing a method to estimate atmospheric light effectively in highlight-dominant scenes.
  • Proposed a highlight-aware dehazing framework based on sky-feature constraints.
  • Implemented real-time processing on ZYNQ7020 using PS-PL co-design.
  • Utilized dark channel for adaptive transmission estimation and radiance recovery.
  • Achieved 19.68 dB PSNR and 0.8196 SSIM on SOTS-Outdoor benchmark.
  • Outperformed classical DCP baseline by 3.62 dB in PSNR.
  • Real-time system reached 108 FPS on ZYNQ7020.

Cite This Study

Zhao et al. (2026) studied this question.

synapsesocial.com/papers/6a1fc616dee9eb8c0dce757ehttps://doi.org/10.1007/s10791-026-10194-x
View Full Paper
Ask AI
Bookmark
Share