基于反射分类与梯度恢复的单幅图像去反射模型  

Single Image Reflection Removal Model Based on Reflection Classifier and Gradient Restorer

作　　者：解庆 张凌峰 马艳春 刘永坚[1,2] XIE Qing;ZHANG Lingfeng;MA Yanchun;LIU Yongjian(Sehool of Computer Sxience and Artificial Intelligence,Wuhan University of Technology,Wuhan 430070,Hubei,China;Engineering Research Center of Imteligent Service Tehmology for Digital Putlishing,Ministry of Eaucation,Wuhan 430070,Hubei.China;School of Management,Wuhan University of Technology,Wuhan 430070,Hubei,China)

机构地区：[1]武汉理工大学计算机与人工智能学院,湖北武汉430070 [2]数字出版智能服务技术教育部工程研究中心,湖北武汉430070 [3]武汉理工大学管理学院,湖北武汉430070

出　　处：《计算机工程》2025年第4期227-238,共12页Computer Engineering

基　　金：国家自然科学基金面上项目(62271360);湖北省重点研发计划项目(2023BAB085)。

摘　　要：单幅图像反射去除是计算机视觉领域的一项重要任务。然而,现有的图像反射去除模型都基于反射污染区域属于模糊型反射这一前提,即反射区域仍然保留原始的图像内容信息。当污染图像中存在光斑反射时,图像原始内容信息完全丢失,导致现有模型无法从光斑区域中提取原始图像的透射层信息,从而使模型失效。针对这一问题,提出一种能够同时去除光斑与模糊反射的新模型,通过自定义的反射分类器和结构恢复器引导模型预测图像透射层的梯度图,并以此作为辅助条件,最终生成纯净的透射层图像。实验结果表明,该模型对不同类别的反射图像均具有较好的泛化性能,在艺术图像唐卡上,模型在结构相似度(SSIM)与峰值信噪比(PSNR)指标上均优于当前最优的反射去除模型,其中SSIM与最优模型相比提升了1.6%,PSNR提升了5.5%。在公共的自然场景数据集上的实验结果也表明该模型与当前最优模型性能相当。Removing single image reflections is an important task in computer vision.However,existing image reflection removal models are based on the premise that reflection pollution areas are fuzzy types,which means the reflection areas retain the original image content information.In the case of spot reflection in a contaminated image,the original content information of the image is completely lost,leading to the failure of existing models in extracting the original image transmission layer information from the spot region.To address this problem,this study proposes a new model that can simultaneously remove spots and fuzzy reflections.By utilizing a self-defined reflection classifier and structure restorer,the model predicts the gradient map of the image transmission layer and uses it as an auxiliary condition to generate an ultimately pure transmission-layer image.Experiments show that our model has a good generalization performance on different categories of reflected images.Experiments on art images,specifically Tangka,demonstrate that our model outperforms the state-of-the-art removal model in terms of Structure Similarity Index Measure(SSIM)and Peak Signal-to-Noise Ratio(PSNR),which increase by 1.6%and 5.5%,respectively.Experiments on public natural scene datasets also indicate that our model is comparable to state-of-the-art models.

关 键 词：单幅图像反射去除 反射分类 图像梯度恢复 生成对抗网络 注意力机制 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]