首页|隧道场景下行人检测DA-Zero-DCE图像增强算法

隧道场景下行人检测DA-Zero-DCE图像增强算法

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
隧道图像受拍摄环境影响,存在光照分布不均、局部遮挡、噪点较多等问题,针对现有图像增强算法在优化过程中的过曝与失真,提出一种隧道图像增强算法DA-Zero-DCE(denoising-attention based zero-reference deep curve estimation).首先,基于 Zero-DCE 模型,使用U-Net改进用于曲线估计的主干网络DCE-Net,并且加入坐标注意力机制来提升对图像局部区域的暗光感知能力.其次,在曲线估计主干网络后加入NAF-Net噪声去除模块,有效抑制Zero-DCE在低光照增强后的噪声.此外,为缓解增强图像的失真与过曝现象,将空间一致性损失函数的4邻域计算方式扩展为8邻域计算方式,增强输出结果平滑度.通过LOL数据集的消融实验,DA-Zero-DCE模型比Zero-DCE模型在增强结果上的PSNR(峰值信噪比)提升约10 dB,SSIM(结构相似性)提升约0.1,验证了模型的有效性和可行性.
A denoising-attention based Zero-DCE for tunnel image enhancement
Tunnel images,affected by the shooting environment,suffer from uneven illumination distribution,local occlusion,and many noises.To address the problems of overexposure and distortion in existing image enhancement algorithms,this paper proposes a tunnel image enhancement algorithm called DA-Zero-DCE(Denoising-Attention based Zero-Reference Deep Curve Estimation).First,based on the Zero-DCE model,the U-Net is employed to improve the backbone network DCE-Net for curve estimation,and a coordinate attention mechanism is added to enhance the dark light perception ability of local image areas.Second,the NAF-Net noise removal module is added after the curve estimation backbone network to effectively suppress the noises after low-light enhancement by Zero-DCE.To offset the distortion and overexposure of the enhanced images,the 4-neighborhood calculation method of the spatial consistency loss function is extended to an 8-neighborhood calculation method,enhancing the smoothness of the outputs.Through the ablation experiment on the LOL dataset,the DA-Zero-DCE model,compared to the Zero-DCE model,improves PSNR by 10 dB and SSIM by 0.1,demonstrating its feasibility and effectiveness.

deep learningconvolutional neural networkcomputer visiontunnel image enhance-ment

周桐、李冬春、田雨聃

展开 >

重庆师范大学计算机与信息科学学院,重庆 401331

重庆工程职业技术学院大数据与物联网学院,重庆 402260

重庆大学 自动化学院,重庆 400044

深度学习 卷积神经网络 计算机视觉 图像增强

中国高校产学研创新基金项目重庆市教委科学技术研究计划项目重庆市教委科学技术研究计划项目重庆英才计划"包干制"项目

2021BCG02002KJZD-K202303405KJQN202303423cstc2022ycjhbgzxm0108

2024

10.3969/j.issn.1674-8425(z).2024.01.014

重庆理工大学学报
重庆理工大学

重庆理工大学学报

CSTPCD北大核心
影响因子：0.567
ISSN：1674-8425
年,卷(期)：2024.38(1)
  • 2