基于Word Embedding的遥感影像检测分割

遥感影像检测分割技术通常需提取影像特征并通过深度学习算法挖掘影像的深层特征来实现.然而传统特征（如颜色特征、纹理特征、空间关系特征等）不能充分描述影像语义信息,而单一结构或串联算法无法充分挖掘影像的深层特征和上下文语义信息.针对上述问题,本文通过词嵌入将空间关系特征映射成实数密集向量,与颜色、纹理特征的结合.其次,本文构建基于注意力机制下图卷积网络和独立循环神经网络的遥感影像检测分割并联算法（Attention Graph Convolution Networks and Independently Recurrent Neural Network,ATGIR）.该算法首先通过注意力机制对结合后的特征进行概率权重分配;然后利用图卷积网络（GCNs）算法对高权重的特征进一步挖掘并生成方向标签,同时使用独立循环神经网络（IndRNN）算法挖掘影像特征中的上下文信息,最后用Sigmoid分类器完成影像检测分割任务.以胡杨林遥感影像检测分割任务为例,我们验证了提出的特征提取方法和ATGIR算法能有效提升胡杨林检测分割任务的性能.     

结合空洞可分离卷积的多尺度医学影像分割算法

传统U-Net语义分割模型在医学影像领域具有广泛的应用,但该模型的准确率受限于单一尺度的预测模式以及上下采样引起的信息丢失。针对上述问题,本文基于U-Net编码—解码架构以及空洞可分离卷积提出了一种高低层级信息丰富的多尺度医学影像语义分割算法,该算法由特征提取网络以及多尺度语义分割预测网络两部分构成。特征提取网络使用空洞可分离卷积和类残差块分别替换原U-Net中上、下采样以及卷积块,在增加感受野的同时使信息得到最大化的保留;提出一个通道注意力机制,强化目标核心特征的表达以及无关背景区域的抑制;在多尺度上挖掘带有图像级全局上下文的卷积特征,进一步提高分割性能。本文在采集的胚胎以及DRIVE数据集上进行仿真实验,其结果表明,与U-Net及其衍生模型相比该方法具有更高的准确率和鲁棒性。     

基于多级特征补偿的遥感图像时空融合方法

在许多地球科学应用中要用到大量的高时空分辨力的地球观测数据。时空图像融合方法为产生高时空分辨力的数据提供了一种可行且经济的解决方案。然而,现有的一些基于学习的方法对于图像深层特征提取能力较弱,对于高分辨力图像细节特征利用度不够。针对这些问题,提出一种基于多级特征补偿的遥感图像时空融合方法。该方法使用2个分支进行多层级的特征补偿,并提出了融合通道注意力机制的残差模块作为网络的基本组成单元,可以将高分辨力输入图像的深层特征更为详尽地提取利用。提出一种基于拉普拉斯算子的边缘损失,在节省预训练计算开销的同时取得了很好的融合效果。使用从山东和广东2个地区采集的Landsat和中分辨力成像光谱仪(MODIS)卫星图像对所提出的方法进行实验评估。实验结果表明,提出的方法在视觉外观和客观指标方面都具有更高质量。     

Single image deblurring with cross-layer feature fusion and consecutive attention

Single image deblurring aims to restore the single blurry image to its sharp counterpart and remains an active topic of enduring interest. Recently, deep Convolutional Neural Network (CNN) based methods have achieved promising performance. However, two primary limitations mainly exist on those CNNs-based image deblurring methods: most of them simply focus on increasing the complexity of the network, and rarely make full use of features extracted by encoder. Meanwhile, most of the methods perform the deblurred image reconstruction immediately after the decoder, and the roles of the decoded features are always underestimated. To address these issues, we propose a single image deblurring method, in which two modules to fuse multiple features learned in encoder (the Cross-layer Feature Fusion (CFF) module) and manipulate the features after decoder (the Consecutive Attention Module (CAM)) are specially designed, respectively. The CFF module is to concatenate different layers of features from encoder to enhance rich structural information to decoder, and the CAM module is able to generate more important and correlated textures to the reconstructed sharp image. Besides, the ranking content loss is employed to further restore more realistic details in the deblurred images. Comprehensive experiments demonstrate that our proposed method can generate less blur and more textures in deblurred image on both synthetic datasets and real-world image examples.     

基于特征金字塔融合表征网络的跨模态哈希方法

随着多模态数据的爆发式增长,跨模态检索作为一种搜索多模态数据的最常用方法,受到越来越多的关注.然而,目前存在的大多数深度学习的方法仅仅采用模型后端最后一个全连接层输出作为模态独有的高层语义表征,忽视了多个层次上不同尺度特征之间的语义相关性,具有一定的局限性.为此,本文提出一种基于特征金字塔融合表征网络的跨模态哈希检索方...     

基于Swin Transformer和混合特征聚合的红外与可见光图像融合方法

红外与可见光图像融合可以生成包含更多信息的图像,比原始图像更符合人类视觉感知也有利于下游任务的进行。传统的基于信号处理的图像融合方法存在泛化能力不强、处理复杂图片融合性能下降等问题。深度学习有很强的特征提取能力,其生成的结果较好,但结果中存在纹理细节信息保存少、图像模糊的问题。针对这一问题,文中提出一种基于多尺度Swin-transformer和注意力机制的红外与可见光图像融合网络模型。Swin-transformer可以在多尺度视角下提取长距离语义信息,注意力机制可以将所提特征中的不重要特征弱化,保留主要信息。此外本文提出了一种新的混合特征聚合模块,针对红外和可见光图像各自的特点分别设计了亮度增强模块和细节保留模块,有效保留更多的纹理细节和红外目标信息。该融合方法包括编码器、特征聚合和解码器三部分。首先,将源图像输入编码器,提取多尺度深度特征;然后,设计特征聚合融合每个尺度的深度特征;最后,采用基于嵌套连接的解码器重构融合后的图像。在公开数据集上的实验结果表明本文提出的方法对比其他先进的方法具有更好的融合性能。其中在客观评价指标中EI、AG、QP、EN、SD指标达到最优。从主观感受上,所提红外和可见光图像融合方法能够使结果中保留更多的边缘细节。     

基于自注意力深度网络的图像超分辨率重建方法

针对现有图像超分辨重建方法难以充分重建图像的细节信息且易出现重建的图像缺乏层次的问题,提出一种基于自注意力深度网络的图像超分辨重建方法。以深度神经网络为基础,通过提取低分辨率图像特征,建立低分辨率图像特征到高分辨率图像特征的非线性映射,重建高分辨率图像。在进行非线性映射时,引入自注意力机制,获取图像中全部像素间的依赖关系,利用图像的全局特征指导图像重建,增强图像层次。在训练深度神经网络时,使用图像像素级损失和感知损失作为损失函数,以强化网络对图像细节信息的重建能力。在3类数据集上的对比测试结果表明,所提方法能够提升图像超分辨重建结果的细节信息,且重建图像的视觉效果更好。     

结合可逆网络的超高分辨率图像的风格迁移

针对处理超高分辨率图像时面临的内存成本和风格迁移过程中过度风格化,提出了一种结合可逆网络的超高分辨率图像的风格迁移方法。该算法采用可逆的Glow模块作为基本单元构建可逆神经网络模块,并将图像分为小块处理;在风格迁移模块中使用具有通道注意力机制的残差模块和缩略图实例化归一化模块(TIN),以保证各模块风格一致;提出基于全局-局部的损失计算方式,能够有效地处理局部的结构特征。实验结果表明,相较于当前通用的各种神经风格迁移网络,所提算法不仅能够避免图像在编码和解码过程中的信息丢失问题,而且能以更低的内存成本实现更优的风格迁移。     

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.     

YOLO-NKLT视觉SLAM回环检测方法

针对存在明显光照变化或遮挡物等室外复杂场景下,现有基于深度学习的视觉即时定位与地图构建(visual simultaneous localization and mapping,视觉SLAM)回环检测方法没有很好地利用图像的语义信息、场景细节且实时性差等问题,本文提出了一种YOLO-NKLT视觉SLAM回环检测方法。采用改进损失函数的YOLOv5网络模型获取具有语义信息的图像特征,构建训练集,对网络重训练,使提取的特征更加适用于复杂场景下的回环检测。为了进一步提高闭环检测的实时性,提出了一种基于非支配排序的KLT降维方法。通过在New College数据集和光照等变化更复杂的Nordland数据集上进行实验,结果表明：室外复杂场景下,相较于其他传统和基于深度学习的方法,所提方法具有更高的鲁棒性,可以取得更佳的准确率和实时性表现。     

融合视觉机制和多尺度特征的小目标检测算法

针对SSD(Single Shot MultiBox Detector)目标检测算法对小目标检测能力不足的问题,提出一种引入视觉机制和多尺度语义信息融合的VFF-SSD(Vision Feature Fusion SSD)改进算法。为了增大浅层网络的感受野提高特征提取能力,首先在SSD浅层特征层中加入视觉机制,然后利用改进PANet(Path Aggregation Network)多尺度特征融合网络与深层特征增强网络得到新的特征层,旨在增强浅层网络的语义信息并加强深层特征的特征表达能力,最后应用注意力机制模块提高对重要信息的学习能力。实验结果表明,在PASCAL VOC2007测试集检测的mAP(Mean Average Precision)值达到81.1%,对数据集中小目标的mAP值较原SSD提高了6.6%。     

特征反馈机制优化的超声图像病灶检测算法

该文提出一种基于特征反馈机制的超声图像病灶检测方法,以实现超声病灶的实时精确定位与检测。所提方法由基于特征反馈机制的特征提取网络和基于分治策略的自适应检测头两部分组成。特征反馈网络通过反馈特征选取和加权融合计算,充分学习超声图像的全局上下文信息和局部低级语义细节以提高局部病灶特征的识别能力。自适应检测头对特征反馈网络所提取的多级特征进行分治预处理,通过将生理先验知识与特征卷积相结合的方式对各级特征分别进行病灶形状和尺度特征的自适应建模,增强检测头对不同大小病灶在多级特征下的检测效果。所提方法在甲状腺超声图像数据集上进行了测试,得到了70.3%的AP,99.0%的AP50和88.4%的AP75,实验结果表明,相较于主流检测算法,所提算法能实现更精准的实时超声图像病灶检测和定位。     

基于空间特征融合的双路径图像去噪网络

深度卷积神经网络(Convolutional Neural Network, CNN)在图像去噪领域受到广泛关注。然而,随着网络深度的增加,大多数深度CNN会出现性能饱和、学习能力下降等问题。提出了一种结合局部和全局特征的双路径去噪网络,将两个不同结构的网络组合后构成一个双路径模型,增加网络的宽度,从而获得更多不同的特征。通过长路径连接融合全局和局部特征,增强层间相关性。注意力机制利用当前阶段引导前一阶段的输入,获得更多的特征。实验结果表明,我们提出的网络模型在Set12和BSD68两个数据集中的峰值信噪比(Peak Signal-to-Noise Ratio, PSNR)分别达到了32.95 dB和31.74dB。同时,主观视觉效果(如图像的边缘等细节)恢复得更好、更清晰。     

基于注意力机制的多分支特征级联图像去雨网络

针对已有去雨网络在不同环境中去雨不彻底和图像细节信息损失严重的问题,本文提出一种基于注意力机制的多分支特征级联图像去雨网络。该模型结合多种注意力机制,形成不同类型的多分支网络,将图像空间细节和上下文特征信息在整体网络中自下而上地进行传递并级联融合,同时在网络分支间构建的阶段注意融合机制,可以减少特征提取过程中图像信息的损失,更大限度地保留特征信息,使图像去雨任务更加高效。实验结果表明,本文算法的客观评价指标优于其他对比算法,主观视觉效果得以有效提升,去雨能力更强,准确性更加突出,能够去除不同密度的雨纹,并且能够更好地保留图像背景中的细节信息。     

Research on image feature extraction and retrieval algorithms based on convolutional neural network

With the rapid development of mobile Internet and digital technology, people are more and more keen to share pictures on social networks, and online pictures have exploded. How to retrieve similar images from large-scale images has always been a hot issue in the field of image retrieval, and the selection of image features largely affects the performance of image retrieval. The Convolutional Neural Networks (CNN), which contains more hidden layers, has more complex network structure and stronger ability of feature learning and expression compared with traditional feature extraction methods. By analyzing the disadvantage that global CNN features cannot effectively describe local details when they act on image retrieval tasks, a strategy of aggregating low-level CNN feature maps to generate local features is proposed. The high-level features of CNN model pay more attention to semantic information, but the low-level features pay more attention to local details. Using the increasingly abstract characteristics of CNN model from low to high. This paper presents a probabilistic semantic retrieval algorithm, proposes a probabilistic semantic hash retrieval method based on CNN, and designs a new end-to-end supervised learning framework, which can simultaneously learn semantic features and hash features to achieve fast image retrieval. Using convolution network, the error rate is reduced to 14.41% in this test set. In three open image libraries, namely Oxford, Holidays and ImageNet, the performance of traditional SIFT-based retrieval algorithms and other CNN-based image retrieval algorithms in tasks are compared and analyzed. The experimental results show that the proposed algorithm is superior to other contrast algorithms in terms of comprehensive retrieval effect and retrieval time.     

Global attention retinex network for low light image enhancement

Most low-light image enhancement methods only adjust the brightness, contrast and noise reduction of low-light images, making it difficult to recover the lost information in darker areas of the image, and even cause color distortion and blurring. To solve the above problems, a global attention-based Retinex network (GARN) for low-light image enhancement is proposed in this paper. We propose a novel global attention module which computes multiple dimensional information in the channel attention module to help facilitate inference learning. Then the global attention module is embedded into different layers of the network to extract richer shallow texture features and deep semantic features. This means that the rich features are more conducive to learning the mapping relationship between low-light images to normal-light images, so that the detail recovery of dark regions is enhanced in low-light images. We also collected a low/normal light image dataset with multiple scenes, in which the images paired as training set can succeed to be applied to low-light image enhancement under different lighting conditions. Experimental results on publicly available datasets show that our method has better effectiveness and generality than the state-of-the-art methods in terms of evaluations metrics such as PSNR, SSIM, NIQE, Entropy.     

Image forgery detection algorithm based on U-shaped detection network

Aiming at the defects of traditional image tampering detection algorithm relying on single image attribute,low applicability and current high time-complexity detection algorithm based on deep learning,an U-shaped detection network image forgery detection algorithm was proposed.Firstly,the multi-stage feature information in the image by using the continuous convolution layers and the max-pooling layers was extracted by U-shaped detection network,and then the obtained feature information to the resolution of the input image through the upsampling operation was restored.At the same time,in order to ensure higher detection accuracy while extracting high-level semantic information of the image,the output features of each stage in U-shaped detection network would be merged with the corresponding output features through the upsampling layer.Further the hidden feature information between tampered and un-tampered regions in the image upon the characteristics of the general network was explored by U-shaped detection network,which could be realized quickly by using its end-to-end network structure and extracting the attributes of strong correlation information among image contexts that could ensure high-precision detection results.Finally,the conditional random field was used to optimize the output of the U-shaped detection network to obtain a more exact detection results.The experimental results show that the proposed algorithm outperforms those traditional forgery detection algorithms based on single image attribute and the current deep learning-based detection algorithm,and has good robustness.     

基于空谱注意力机制及预激活残差网络的高光谱图像分类算法

面向高光谱图像分类的许多深度学习算法中,由于提取的空谱特征表示鉴别性不足,其模型的分类性能有待提高。针对该问题,本文提出了一种基于空谱注意力机制及预激活残差网络的高光谱图像分类算法。首先,设计了基于空谱注意力机制的空谱特征提取模块,对空谱特征进行重校准,为空谱特征在后续联合学习时能专注于更具辨别力的通道和空间位置提供保证;其次,设计了基于预激活残差网络的空谱特征联合学习模块,其中预激活残差网络改进了原始残差构建块的网络结构,从而能在利用注意力机制重校准的空谱特征的联合学习时捕获更具鉴别性的深层空谱特征,以提高分类器的分类性能。实验结果表明,和已有的一些高光谱图像分类算法相比,所提出的算法的分类准确率更高,表明该算法能有效地获得判别能力更强的空谱特征表示。      

基于一种视觉注意力机制的图像描述方法

为了提高由图像生成文字描述的准确率,文中提出了一种基于传统的编码解码框架,分别在编码端和解码端融入视觉注意力机制的方法,即在编码端加入空间注意力机制和图像通道级注意力机制相结合的方法。在解码端运用自适应视觉注意力机制的方法,即在传统的解码端上加入一个额外的“视觉哨兵”模块。文中提出的方法在生成文字描述的过程中自动决定是依赖图像特征还是依赖语义特征,并传递给相应的注意力机制。实验证明,相比较单一的视觉注意力机制,文中方法取得了较高的图像描述语句的正确率,具有更好的图像描述性能。     

融入残差注意力机制的DeepLabV3+图像拼接篡改取证网络

针对现有图像拼接检测网络模型存在边缘信息关注度不够、像素级精准定位效果不够好等问题,提出一种融入残差注意力机制的DeepLabV3+图像拼接篡改取证方法,该方法利用编-解码结构实现像素级图像的拼接篡改定位。在编码阶段,将高效注意力模块融入ResNet101的残差模块中,通过残差模块的堆叠以减小不重要的特征比重,凸显拼接篡改痕迹;其次,利用带有空洞卷积的空间金字塔池化模块进行多尺度特征提取,将得到的特征图进行拼接后通过空间和通道注意力机制进行语义信息建模。在解码阶段,通过融合多尺度的浅层和深层图像特征提升图像的拼接伪造区域的定位精度。实验结果表明,在CASIA 1.0、COLUMBIA和CARVALHO数据集上的拼接篡改定位精度分别达到了0.761、0.742和0.745,所提方法的图像拼接伪造区域定位性能优于一些现有的方法,同时该方法对JPEG压缩也具有更好的鲁棒性。