基于多任务和反馈学习的阴影检测方法

自然场景图像中的阴影是直接或间接承接场景信息的载体,可以有效反映场景的环境信息,因此,阴影检测是图像处理领域中的一项重要研究任务。针对阴影检测任务存在的漏检、误检等问题,提出了一种应用于阴影检测任务的网络。引入多任务特征学习机制,利用阴影的边缘和数量信息分别实现特征学习的细节和全局约束,有效地定位阴影区域。此外还利用注意力机制增强特征信息,结合反馈机制通过迭代反馈的方式增强模型对阴影区域特征的学习,提高了模型性能,使模型能够准确区分阴影区域和非阴影区域。实验结果表明,该网络模型与其他经典方法相比在SBU数据集和ISTD数据集上取得了最低的平衡错误率指标,可以更加准确地检测单幅图像中存在的阴影区域且边缘划分准确,检测结果令人满意。     

基于多级残差融合的复杂纹理光场图像深度估计

光场的深度信息可以通过深度学习的光场深度估计算法计算,在图像视差、光场图像边缘以及光场图像的复杂纹理区域,获取高精度深度值仍然具有一定局限性。本文提出了一种用于光场图像深度估计的多级残差融合网络,通过组合残差模块提取多层次的残差特征,在保持网络深度的同时提升了网络对特征的表征能力。利用多级残差融合模块对多层次的残差特征进行融合,以获得包含浅层纹理信息和深层语义信息的融合特征。利用本文方法对HCI4D光场数据集进行处理,图像深度估计的均方误差指标达到1.471,不良像素率指标达到4.208,该实验结果表明本文方法在处理具有复杂遮挡的光场图像区域方面具有良好的处理效果。     

基于深度阴影特征增强的任意至任意重光照

任意至任意重光照利用隐含在引导图像中的光照重新照明源图像。现有的任意至任意重光照方法由于采用端到端的学习方式,导致阴影特征与色温特征高度耦合,进一步影响了阴影生成的准确性。为此,本文提出了一个基于深度阴影特征增强的任意至任意重光照方法。该方法的关键是设计一个额外的阴影解码器,从隐式表征中直接生成对应的阴影图像。同时,为了充分利用学习到的阴影特征,我们引入一个基于注意力机制的特征融合模块,实现重光照特征与阴影特征的自适应融合。另外,我们实验性地发现,利用多项式核函数把源图像映射到高维特征后,再作为网络输入,能进一步提升性能。在VIDIT数据集上的实验表明了本文所提方法的有效性。      

基于全卷积网络的语义显著性区域检测方法研究

基于底层视觉特征和先验知识的显著性区域检测算法难以检测一些复杂的显著性目标,人的视觉系统能分辨这些目标是由于其中包含丰富的语义知识.本文构建了一个基于全卷积结构的语义显著性区域检测网络,用数据驱动的方式构建从图像底层特征到人类语义认知的映射,提取语义显著性区域.针对网络提取的语义显著性区域的缺点,本文进一步引入颜色信息、目标边界信息、空间一致性信息获得准确的超像素级前景和背景概率.最后提出一个优化模型融合前景和背景概率信息、语义信息、空间一致性信息得到最终的显著性区域图.在6个数据集上与15种最新算法的比较实验证明了本文算法的有效性和鲁棒性.     

多层级特征优化融合的遥感图像分割网络

为了从高分辨率遥感图像中准确地分割出地物目标，提出了一种多级特征优化融合的遥感图像分割网络（MRFNet），着重将特征提取骨架网络中不同层级的特征图进行融合，通过融合网络特征图中不同种类的信息来对输出特征图信息进行合理有效的提取和分析。同时使用了逐层的多尺度编码解码模块来细化与高层特征图进行融合的浅层特征图，将不同种类的信息经过优化以后汇聚到高层特征图。然后采用空洞卷积金字塔对高层特征图提取不同感受野的信息，优化了语义分割的输出特征图。通过在ISPRS Vaihingen数据集上进行实验，该网络算法的总体精度达到了90.34%，与经典语义分割网络相比，有效提升了遥感图像目标的检测精度。同时为了证明算法的泛化性，在ISPRS Potsdam数据集上进行了泛化实验取得了91.47%的总体精度，证明了该算法的有效性。     

基于特征融合与自注意力机制的图像语义分割算法

提出了一种基于特征融合与自注意力机制的图像语义分割方法,设计了特征融合模块、自注意力模块、增强模块、全局空间信息融合模块和损失函数。特征融合模块融合多个图像的所有组件,通过自注意力机制来执行。自注意力模块从而有效地捕获远程上下文信息。增强模块旨在增强输入图像以获得更多样化的特征。全局空间信息注意模块相对于图像尺寸只有线性的复杂度,能够带来显著的提升效果。利用损失函数,对模型进行优化,将每个像素的分类结果优化到最接近真实值。实验结果表明,所提出的方法可以显著提高PASCAL VOC 2012数据集、COCO-Stuff 10K数据集和ISIC 2018数据集这3个数据集的性能,并在3个数据集上进行了验证,实验还通过对自注意力、推理速度和消融实验进行比较,验证了本文方法的优越性。     

一种多分辨率特征提取红外图像语义分割算法

针对现有图像语义分割算法在对低分辨率红外图像进行分割时存在准确率不高的问题,提出了一种多分辨率特征提取算法。该算法以DeepLabv3+为基准网络,添加了一组对偶分辨率模块,该模块包含低分辨率分支和高分辨率分支,以进一步聚合红外图像特征。低分辨率分支采用GPU友好的注意力模块捕获高层全局上下文信息,同时引入一个多轴门控感知机模块并行提取红外图像局部信息和全局信息;高分辨率分支采用跨分辨率注意力模块将低分辨率分支上学习到的全局特征传播扩散到高分辨率分支上以获取更强的语义信息。实验结果表明,该算法在数据集DNDS和MSRS上的分割精度优于现有语义分割算法,证明了提出算法的有效性。     

面向自然场景分类的贝叶斯网络局部语义建模方法

本文提出了一种基于贝叶斯网络的局部语义建模方法.网络结构涵盖了区域邻域的方向特性和区域语义之间的邻接关系.基于这种局部语义模型,建立了场景图像的语义表述,实现自然场景分类.通过对已标注集的图像样本集的学习训练,获得贝叶斯刚络的参数.对于待分类的图像,利用该模型融合区域的特征及其邻接区域的信息,推理得到区域的语义概率;并通过网络迭代收敛得到整幅图像的区域语义标记和语义概率;最后在此基础上形成图像的全局描述,实现场景分类.该方法利用了场景内部对象之间的上下文关系,弥补了仅利用底层特征进行局部语义建模的不足.通过在六类自然场景图像数据集上的实验表明,本文所提的局部语义建模和图像描述方法是有效的.     

基于Ghost模块的改进YOLOv5目标检测算法

现有以YOLOv5为代表的目标检测技术,存在骨干网络对特征提取不充分以及颈部层未高效融合浅层位置信息和深层高级语义信息等问题,这会导致检测精度较低,小目标误检、漏检。针对此问题,从兼顾实时性与检测精度出发,对YOLOv5进行改进,提出一种改进网络YOLOv5-CBGhost。首先在骨干网络中引入Ghost模块对模型进行轻量化处理,引入CA模块来更好地获得全局感受野,提高模型获取目标位置的准确度;然后借鉴双向加权特征金字塔网络,对原PAN结构进行改进,有效减少了特征冗余以及参数量,并通过跨层加权连接融合更多特征,提高了模型的目标检测精度;最后,增加多检测头以获取图片更丰富的高层语义信息,有效增加了检测精度。通过在PASCAL VOC2007+2012数据集上实验,YOLOv5-CBGhost的目标精度达到81.8%,相较于YOLOv5s,提高了3.0%,计算量减少42.5%,模型大小减少3.5%。     

基于Double-Head的雾天图像目标检测

雾天环境下的图像对比度低,图像中的目标较为模糊并且其特征提取存在一定难度。现有的目标检测方法对于雾天图像的检测准确率偏低。针对上述问题,本文在Double-Head框架上基于图像的特征提取部分和预测头部进行改进。首先,在提取的深层特征图上添加通道和空间双维度的复合注意力机制,提高网络关注显著目标的能力;其次,将原始图像经过改进的暗通道先验以及处理后得到的先验矩阵和特征图进一步融合,获取更全面的雾天图像特征信息;最后,在预测头部引入可分离卷积,使用解耦合预测头对目标进行最终的分类和回归。该方法在RTTS数据集上的mAP为49.37%,在合成数据集S-KITTI和S-COCOval数据集上的AP值分别为66.7%和57.7%。与其他主流算法相比,本文算法具有更高的目标检测精度。     

Global and local information aggregation network for edge-aware salient object detection

Aggregation of local and global contextual information by exploiting multi-level features in a fully convolutional network is a challenge for the pixel-wise salient object detection task. Most existing methods still suffer from inaccurate salient regions and blurry boundaries. In this paper, we propose a novel edge-aware global and local information aggregation network (GLNet) to fully exploit the integration of side-output local features and global contextual information and utilization of contour information of salient objects. The global guidance module (GGM) is proposed to learn discriminative multi-level information with the direct guidance of global semantic knowledge for more accurate saliency prediction. Specifically, the GGM consists of two key components, where the global feature discrimination module exploits the inter-channel relationship of global semantic features to boost representation power, and the local feature discrimination module enables different side-output local features to selectively learn informative locations by fusing with global attentive features. Besides, we propose an edge-aware aggregation module (EAM) to employ the correlation between salient edge information and salient object information for generating estimated saliency maps with explicit boundaries. We evaluate our proposed GLNet on six widely-used saliency detection benchmark datasets by comparing with 17 state-of-the-art methods. Experimental results show the effectiveness and superiority of our proposed method on all the six benchmark datasets.     

Structure-prior deep neural network for lane detection

Lane detection is an important task of road environment perception for autonomous driving. Deep learning methods based on semantic segmentation have been successfully applied to lane detection, but they require considerable computational cost for high complexity. The lane detection is treated as a particular semantic segmentation task due to the prior structural information of lane markings which have long continuous shape. Most traditional CNN are designed for the representation learning of semantic information, while this prior structural information is not fully exploited. In this paper, we propose a recurrent slice convolution module (called RSCM) to exploit the prior structural information of lane markings. The proposed RSCM is a special recurrent network structure with several slice convolution units (called SCU). The RSCM could obtain stronger semantic representation through the propagation of the prior structural information in SCU. Furthermore, we design a distance loss in consideration of the prior structure of lane markings. The lane detection network can be trained more steadily via the overall loss function formed by combining segmentation loss with the distance loss. The experimental results show the effectiveness of our method. We achieve excellent computation efficiency while keeping decent detection quality on lane detection benchmarks and the computational cost of our method is much lower than the state-of-the-art methods.     

Single image shadow detection via uncertainty analysis and GCN-based refinement strategy

Learning-based shadow detection methods have achieved an impressive performance, while these works still struggle on complex scenes, especially ambiguous soft shadows. To tackle this issue, this work proposes an efficient shadow detection network (ESDNet) and then applies uncertainty analysis and graph convolutional networks for detection refinement. Specifically, we first aggregate global information from high-level features and harvest shadow details in low-level features for obtaining an initial prediction. Secondly, we analyze the uncertainty of our ESDNet for an input shadow image and then take its intensity, expectation, and entropy into account to formulate a semi-supervised graph learning problem. Finally, we solve this problem by training a graph convolution network to obtain the refined detection result for every training image. To evaluate our method, we conduct extensive experiments on several benchmark datasets, i.e., SBU, UCF, ISTD, and even on soft shadow scenes. Experimental results demonstrate that our strategy can improve shadow detection performance by suppressing the uncertainties of false positive and false negative regions, achieving state-of-the-art results.     

基于双分支特征融合的无锚框目标检测算法

针对无锚框目标检测算法CenterNet中,目标特征利用程度不高、检测结果不够准确的问题,该文提出一种双分支特征融合的改进算法。在算法中,一个分支包含了特征金字塔增强模块和特征融合模块,以对主干网络输出的多层特征进行融合处理。同时,为利用更多的高级语义信息,在另一个分支中仅对主干网络的最后一层特征进行上采样。其次,对主干网络添加了基于频率的通道注意力机制,以增强特征提取能力。最后,采用拼接和卷积操作对两个分支的特征进行融合。实验结果表明,在公开数据集PASCAL VOC上的检测精度为82.3%,比CenterNet算法提高了3.6%,在KITTI数据集上精度领先其6%,检测速度均满足实时性要求。该文提出的双分支特征融合方法将不同层的特征进行处理,更好地利用浅层特征中的空间信息和深层特征中的语义信息,提升了算法的检测性能。     

SDRNet: An end-to-end shadow detection and removal network

Image shadow detection and removal can effectively recover image information lost in the image due to the existence of shadows, which helps improve the accuracy of object detection, segmentation and tracking. Thus, aiming at the problem of the scale of the shadow in the image, and the inconsistency of the shadowed area with the original non-shadowed area after the shadow is removed, the multi-scale and global feature (MSGF) is used in the proposed method, combined with the non-local network and dense dilated convolution pyramid pooling network. Besides, aiming at the problem of inaccurate detection of weak shadows and complicated shape shadows in existing methods, the direction feature (DF) module is adopted to enhance the features of the shadow areas, thereby improving shadow segmentation accuracy. Based on the above two methods, an end-to-end shadow detection and removal network SDRNet is proposed. SDRNet completes the task of sharing two feature heights in a unified network without adding additional calculations. Experimental results on the two public datasets ISDT and SBU demonstrate that the proposed method achieves more than 10% improvement in the BER index for shadow detection and the RMSE index for shadow removal, which proves that the proposed SDRNet based on the MSGF module and DF module can achieve the best results compared with other existing methods.     

Cross-level reinforced attention network for person re-identification

Attention mechanism is a simple and effective method to enhance discriminative performance of person re-identification (Re-ID). Most of previous attention-based works have difficulty in eliminating the negative effects of meaningless information. In this paper, a universal module, named Cross-level Reinforced Attention (CLRA), is proposed to alleviate this issue. Firstly, we fuse features of different semantic levels using adaptive weights. The fused features, containing richer spatial and semantic information, can better guide the generation of subsequent attention module. Then, we combine hard and soft attention to improve the ability to extract important information in spatial and channel domains. Through the CLRA, the network can aggregate and propagate more discriminative semantic information. Finally, we integrate the CLRA with Harmonious Attention CNN (HA-CNN) and form a novel Cross-level Reinforced Attention CNN (CLRA-CNN) to optimize person Re-ID. Experiment results on several public benchmarks show that the proposed method achieves state-of-the-art performance.     

Multi-scale gradient attention guidance and adaptive style fusion for image inpainting

Image inpainting aims to fill in the missing regions of damaged images with plausible content. Existing inpainting methods tend to produce ambiguous artifacts and implausible structures. To address the above issues, our method aims to fully utilize the information of known regions to provide style and structural guidance for missing regions. Specifically, the Adaptive Style Fusion (ASF) module reduces artifacts by transferring visual style features from known regions to missing regions. The Gradient Attention Guidance (GAG) module generates accurate structures by aggregating semantic information along gradient boundary regions. In addition, the Multi-scale Attentional Feature Extraction (MAFE) module extracts global contextual information and enhances the representation of image features. The sufficient experimental results on the three datasets demonstrate that our proposed method has superior performance in terms of visual plausibility and structural consistency compared to state-of-the-art inpainting methods.     

多尺度语义信息融合的目标检测

针对当前目标检测算法对小目标及密集目标检测效果差的问题,该文在融合多种特征和增强浅层特征表征能力的基础上提出了浅层特征增强网络(SEFN),首先将特征提取网络VGG16中Conv4_3层和Conv5_3层提取的特征进行融合形成基础融合特征;然后将基础融合特征输入到小型的多尺度语义信息融合模块中,得到具有丰富上下文信息和空间细节信息的语义特征,同时把语义特征和基础融合特征经过特征重利用模块获得浅层增强特征;最后基于浅层增强特征进行一系列卷积获取多个不同尺度的特征,并输入各检测分支进行检测,利用非极大值抑制算法实现最终的检测结果。在PASCAL VOC2007和MS COCO2014数据集上进行测试,模型的平均精度均值分别为81.2%和33.7%,相对于经典的单极多盒检测器(SSD)算法,分别提高了2.7%和4.9%;此外,该文方法在检测小目标和密集目标场景上,检测精度和召回率都有显著提升。实验结果表明该文算法采用特征金字塔结构增强了浅层特征的语义信息,并利用特征重利用模块有效保留了浅层的细节信息用于检测,增强了模型对小目标和密集目标的检测效果。     

基于双重注意力和多尺度特征融合的场景文本检测算法

本文提出了一种场景文本检测方法,用于应对复杂自然场景中文本检测的挑战。该方法采用了双重注意力和多尺度特征融合的策略,通过双重注意力融合机制增强了文本特征通道之间的关联性,提升了整体检测性能。在考虑到深层特征图上下采样可能引发的语义信息损失的基础上,提出了空洞卷积多尺度特征融合金字塔(dilated convolution multi-scale feature fusion pyramid structure, MFPN),它采用双融合机制来增强语义特征,有助于加强语义特征,克服尺度变化的影响。针对不同密度信息融合引发的语义冲突和多尺度特征表达受限问题,创新性地引入了多尺度特征融合模块(multi-scale feature fusion module, MFFM)。此外,针对容易被冲突信息掩盖的小文本问题,引入了特征细化模块(feature refinement module, FRM)。实验表明,本文的方法对复杂场景中文本检测有效,其F值在CTW1500、ICDAR2015和Total-Text 3个数据集上分别达到了85.6%、87.1%和86.3%。