Scholarlink

<p>Underwater images are frequently subject to degradation resulting from challenging environmental conditions, including limited visibility, low illumination, and pronounced light attenuation. This paper presents an effective enhancement approach tailored for underwater image data. The proposed algorithm synergistically integrates Channel Adaptive Color Restoration (CACR), the Zero Approximation Principle (ZAP), and wavelet decomposition techniques. Initially, a modified CACR - ZAP strategy is employed for comprehensive color correction: CACR normalizes RGB channels via adaptive thresholding to achieve balanced color distribution, while ZAP operates in the CIELAB color space to refine chromatic deviation and dynamic range. Subsequently, wavelet decomposition is utilized to separate the low frequency and high frequency of the image and contrast enhancement is performed. Finally, the enhanced LAB space image is converted to RGB, and adaptive histogram equalization is applied for further enhancement. The combined application of CACR, ZAP, and wavelet-based methods demonstrates significant improvements in restoring and enhancing the overall quality of degraded underwater images. Using a dataset of 3473 underwater images, the results strongly support the proposed model, as it achieves the lowest PIQE value of 18.9089 with an 18.57% improvement, and also maintains consistent improvement across the UIQM (3.2845), AG (3.5911), IE (7.6728), and UCIQE (0.8393) metrics.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Color correction and contrast enhancement of underwater images

  • 期刊论文

  • 发表日期：

    2026年3月14日

查看全文
  • Sabareesan Srinivasan,
  • Sakthivel Murugan Santhanam,
  • Muthumeenakshi K

摘要

Underwater images are frequently subject to degradation resulting from challenging environmental conditions, including limited visibility, low illumination, and pronounced light attenuation. This paper presents an effective enhancement approach tailored for underwater image data. The proposed algorithm synergistically integrates Channel Adaptive Color Restoration (CACR), the Zero Approximation Principle (ZAP), and wavelet decomposition techniques. Initially, a modified CACR - ZAP strategy is employed for comprehensive color correction: CACR normalizes RGB channels via adaptive thresholding to achieve balanced color distribution, while ZAP operates in the CIELAB color space to refine chromatic deviation and dynamic range. Subsequently, wavelet decomposition is utilized to separate the low frequency and high frequency of the image and contrast enhancement is performed. Finally, the enhanced LAB space image is converted to RGB, and adaptive histogram equalization is applied for further enhancement. The combined application of CACR, ZAP, and wavelet-based methods demonstrates significant improvements in restoring and enhancing the overall quality of degraded underwater images. Using a dataset of 3473 underwater images, the results strongly support the proposed model, as it achieves the lowest PIQE value of 18.9089 with an 18.57% improvement, and also maintains consistent improvement across the UIQM (3.2845), AG (3.5911), IE (7.6728), and UCIQE (0.8393) metrics.