Scholarlink

Low-resolution (LR) images can only be obtained by long-distance shooting with low light because of the limitations of physical devices. A high-quality detached image is best obtained with optical lenses, which are fairly expensive and bulky. Enhancing low-light Night photography, video surveillance, and remote sensing using Super-Resolution (SR) is essential. We propose an architecture using DenseNet with Skip-connections to transform images from low to high resolution. A dataset containing images captured from various cameras is used for the analysis. In this paper, we used the DPED (DSLR-Photo Enhancement Dataset). Convolution neural networks (CNN) generate results that can be applied to any camera model. As a result of our architecture, we achieved good PSNR and SSIM results.

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

Quality Image Enhancement Using Convolutional Neural Network from Low-Resolution Camera Image

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hassanain K. Alrammahi,
  • Mohammed Brayyich

摘要

Low-resolution (LR) images can only be obtained by long-distance shooting with low light because of the limitations of physical devices. A high-quality detached image is best obtained with optical lenses, which are fairly expensive and bulky. Enhancing low-light Night photography, video surveillance, and remote sensing using Super-Resolution (SR) is essential. We propose an architecture using DenseNet with Skip-connections to transform images from low to high resolution. A dataset containing images captured from various cameras is used for the analysis. In this paper, we used the DPED (DSLR-Photo Enhancement Dataset). Convolution neural networks (CNN) generate results that can be applied to any camera model. As a result of our architecture, we achieved good PSNR and SSIM results.