Related Experiment Video
Updated: Jun 13, 2026
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
DDCATNet: Effective Deep Learning-Based Illumination Color Cast Estimation Approach for Achieving Computational Color
1School of Dentistry, Advanced Dental Device Development Institute, Kyungpook National University, Jung-gu, Daegu 41940, Republic of Korea.
Sensors (Basel, Switzerland)
|June 12, 2026
Summary
This study introduces a new deep learning network, DDCATNet, for more accurate illuminant estimation in computational color constancy (CCC). The novel approach enhances image color accuracy under various lighting conditions, outperforming existing methods.
Area of Science:
- Computer Vision
- Image Processing
- Artificial Intelligence
Background:
- Digital camera sensors struggle to distinguish object colors from illuminant colors.
- Computational Color Constancy (CCC) models human visual perception for accurate color under varying lights.
- Accurate scene illuminant estimation is crucial for effective CCC.
Purpose of the Study:
- To present a novel learning-based approach for illumination color cast estimation in human visual perception-based CCC.
- To improve the precision of illuminant estimation beyond traditional and state-of-the-art methods.
- To enhance the accuracy of object color determination under diverse illuminant conditions.
Main Methods:
- Proposed the Dense Dual Connection Aggregated Transform Network (DDCATNet) architecture.
- Incorporated dense skip connections to integrate channel and spatial features while preserving long-term dependencies.
- Utilized dense channel-wise attention (CA) and spatial attention (SA) blocks within a gate mechanism (GM) for adaptive feature fusion.
Main Results:
- DDCATNet demonstrated significantly enhanced precision in illuminant estimation.
- The proposed approach outperformed state-of-the-art methods on various datasets.
- Validated efficacy, generalization capabilities, and camera-invariance across diverse single- and multi-illuminant scenarios.
Conclusions:
- The DDCATNet architecture offers a robust solution for accurate illuminant estimation in CCC.
- The novel integration of dense dual connections and attention mechanisms improves feature extraction and information preservation.
- The approach shows strong potential for real-world applications requiring reliable color constancy.
Related Concept Videos
01:24Color Vision
Color perception begins in the retina, the light-sensitive layer at the back of the eye. Two main theories explain how colors are seen: the trichromatic theory and the opponent-process theory. The trichromatic theory, proposed by Thomas Young in 1802 and extended by Hermann von Helmholtz in 1852, suggests that color vision is based on three types of cone receptors in the retina. These cones are sensitive to different but overlapping ranges of wavelengths corresponding to red, blue, and green.
01:12Perceptual Constancy
Perceptual constancy is the ability to recognize that objects remain consistent and unchanged even when their appearance varies due to changes in sensory input. There are four main types of perceptual constancy: size constancy, shape constancy, color constancy, and brightness constancy.
Size constancy is the recognition that an object remains the same size, even when its image on the retina changes. For instance, a bus is perceived to be large enough to carry people, even if it looks tiny from...
Size constancy is the recognition that an object remains the same size, even when its image on the retina changes. For instance, a bus is perceived to be large enough to carry people, even if it looks tiny from...