<p>Images captured in low-light environments often suffer from poor visibility and low contrast. Although numerous enhancement techniques have been proposed, most existing methods, while improving brightness and contrast, tend to lose fine details or introduce unnatural artifacts caused by over-enhancement. To address these limitations, this paper proposes a novel image enhancement method that integrates dark region adjustment, image sharpening, and contrast enhancement. The method leverages parametric S-shaped functions: an inverse S-shaped function is first applied to selectively enhance dark regions, followed by multi-scale enhancement with standard S-shaped functions to improve sharpness and contrast while preserving details. In addition, a hue-preserving process is incorporated to maintain natural color appearance. Extensive qualitative and quantitative evaluations demonstrate that the proposed method effectively overcomes the limitations of conventional methods.</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

Low-light Image Enhancement Using Parametric S-Shaped Functions

  • Manaka Yamaguchi,
  • Ayaka Fujita,
  • Takanori Koga,
  • Noriaki Suetake

摘要

Images captured in low-light environments often suffer from poor visibility and low contrast. Although numerous enhancement techniques have been proposed, most existing methods, while improving brightness and contrast, tend to lose fine details or introduce unnatural artifacts caused by over-enhancement. To address these limitations, this paper proposes a novel image enhancement method that integrates dark region adjustment, image sharpening, and contrast enhancement. The method leverages parametric S-shaped functions: an inverse S-shaped function is first applied to selectively enhance dark regions, followed by multi-scale enhancement with standard S-shaped functions to improve sharpness and contrast while preserving details. In addition, a hue-preserving process is incorporated to maintain natural color appearance. Extensive qualitative and quantitative evaluations demonstrate that the proposed method effectively overcomes the limitations of conventional methods.