Saturated Reflection Detection for Reflection Removal Based on Convolutional Neural Network

Taichi Yoshida, Isana FUNAHASHI, Naoki Yamashita, Masaaki Ikehara

Research output: Contribution to journalArticlepeer-review

Abstract

Single image reflection removal is a technique that removes undesirable reflections, which occur due to glass, from images. Various methods of reflection removal have been proposed, but unfortunately, they usually fail to remove reflections particularly with very high pixel values. In this paper, we define these saturated reflections and their characteristics, as well as discuss and propose a removal system. The proposed system detects areas of saturated reflections based on our proposed model of convolutional neural networks and restores them by a conventional method of image estimation. In our experiments, the proposed system shows better peak-signal-to-noise ratio scores and perceptual quality than conventional methods of reflection removal.

Original languageEnglish
JournalIEEE Access
DOIs
Publication statusAccepted/In press - 2022

Keywords

  • Convolution
  • convolutional neural network
  • Convolutional neural networks
  • Estimation
  • Glass
  • image inpainting
  • Laplace equations
  • Licenses
  • Light sources
  • reflection detection
  • Single image reflection removal

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Fingerprint

Dive into the research topics of 'Saturated Reflection Detection for Reflection Removal Based on Convolutional Neural Network'. Together they form a unique fingerprint.

Cite this