基于视觉注意机制的遥感图像显著性目标检测

显著性目标检测是遥感图像处理的重要研究领域,传统的方法通过逐个像素点的计算来实现目标检测,难以满足遥感图像大面积实时处理的要求。将视觉注意机制应用到遥感图像的显著性目标检测中,在训练阶段,将所有的目标融合成目标类,所有的背景融合成背景类,目标类的显著性均值与背景类的显著性均值的比值得到一个权重向量;在检测阶段,所有的特征图乘以权重向量得到自顶向下的显著性图;自顶向下和自底向上的显著性图融合生成全局显著性图。实验结果表明当目标和背景不是总成对出现时,该方法的检测结果优于Navalpakkam模型和Frintrop模型的检测结果。     

Image Style Transfering Based on StarGAN and Class Encoder

The image style transfer technology has been integrated into people''s lives and is widely used in practical scenarios such as artistic images, photo to cartoon, image coloring, filter processing, and occlusion removal, which bears important research significance and application value. StarGAN is a generative adversarial network framework used in recent years for multi-domain image style transfer, which extracts features through simple down-sampling and then generates images through up-sampling. However, the background color information and detailed features of characters'' faces in the generated images are greatly different from those in the input images. In this paper, the network structure of StarGAN is improved, and a UE-StarGAN model for image style transfer is proposed by introducing U-Net and edge-promoting adversarial loss function. At the same time, the class encoder is introduced into the generator of the UE-StarGAN model, and an image style transfer model fusing class encoder based on a small sample size is designed to realize the image style transfer with a small sample size. The experimental results reveal that the model can extract more detailed features and has some advantages in the case of a small sample size. The images obtained by applying the image style transfer based on the proposed model are improved in both qualitative and quantitative analyses, which verifies the effectiveness of the proposed model.     

类别语义相似性监督的小样本图像识别

目的 现有的深度学习模型往往需要大规模的训练数据,而小样本分类旨在识别只有少量带标签样本的目标类别。作为目前小样本学习的主流方法,基于度量的元学习方法在训练阶段大多没有使用小样本目标类的样本,导致这些模型的特征表示不能很好地泛化到目标类。为了提高基于元学习的小样本图像识别方法的泛化能力,本文提出了基于类别语义相似性监督的小样本图像识别方法。方法 采用经典的词嵌入模型GloVe（global vectors for word representation）学习得到图像数据集每个类别英文名称的词嵌入向量,利用类别词嵌入向量之间的余弦距离表示类别语义相似度。通过把类别之间的语义相关性作为先验知识进行整合,在模型训练阶段引入类别之间的语义相似性度量作为额外的监督信息,训练一个更具类别样本特征约束能力和泛化能力的特征表示。结果 在miniImageNet和tieredImageNet两个小样本学习基准数据集上进行了大量实验,验证提出方法的有效性。结果显示在miniImageNet数据集5-way 1-shot和5-way 5-shot设置上,提出的方法相比原型网络（prototypical networks）分类准确率分别提高1.9%和0.32%;在tieredImageNet数据集5-way 1-shot设置上,分类准确率相比原型网络提高0.33%。结论 提出基于类别语义相似性监督的小样本图像识别模型,提高小样本学习方法的泛化能力,提高小样本图像识别的准确率。     

基于未知类语义约束自编码的零样本图像分类

为缓解传统零样本图像分类模型中存在的领域偏移问题,提出一种基于未知类语义约束自编码的零样本图像分类模型.首先,利用预训练的ResNet101网络提取所有已知类和未知类图像的视觉特征;其次,通过编码器将提取的图像深度视觉特征从视觉空间映射到语义空间;然后,通过解码器将映射后得到的语义向量重构为视觉特征向量,在语义自编码器的训练过程中,利用未知类图像的聚类视觉中心和未知类语义类原型的分布对齐施加约束,以缓解领域偏移问题;最后,基于经编码器预测得到的测试图像语义向量和各测试类语义类原型之间的相似性,采用最近邻算法实现零样本图像分类.在AwA2和CUB数据集上的实验结果表明,所提出模型具有较高的分类准确度.     

Periodicity, directionality, and randomness: Wold features forimage modeling and retrieval

One of the fundamental challenges in pattern recognition is choosing a set of features appropriate to a class of problems. In applications such as database retrieval, it is important that image features used in pattern comparison provide good measures of image perceptual similarities. We present an image model with a new set of features that address the challenge of perceptual similarity. The model is based on the 2D Wold decomposition of homogeneous random fields. The three resulting mutually orthogonal subfields have perceptual properties which can be described as “periodicity,” “directionality,” and “randomness,” approximating what are indicated to be the three most important dimensions of human texture perception. The method presented improves upon earlier Wold-based models in its tolerance to a variety of local inhomogeneities which arise in natural textures and its invariance under image transformation such as rotation. An image retrieval algorithm based on the new texture model is presented. Different types of image features are aggregated for similarity comparison by using a Bayesian probabilistic approach. The, effectiveness of the Wold model at retrieving perceptually similar natural textures is demonstrated in comparison to that of two other well-known pattern recognition methods. The Wold model appears to offer a perceptually more satisfying measure of pattern similarity while exceeding the performance of these other methods by traditional pattern recognition criteria. Examples of natural scene Wold texture modeling are also presented     

基于独立关键子块和三角树的快速图象检索新方法

由于图象存储数据量非常大,因此提取图象特征和检索极为耗时.为了提高图象检索效率,将文本检索中的有效检索方法(基于关键字频率与关键字逆文档频率乘积的索引模型)结合三角树索引机制应用到基于内容的图象检索,提出了一种基于独立关键子块和三角树的快速图象检索新方法.该方法首先用独立分量分析将样本图象子块中的直方图特征映射到色彩概念空间来得到类似于文本中关键字的独立关键子块;然后再用训练好的模糊支持向量机去识别每幅图象中所包含的独立关键子块,由于独立分量分析能够使特征彼此保持高阶独立性,因此该方法与主成分分析方法对比,具有较高检索效率;最后,再通过构造三角树来来为图象数据库建立分层索引结构,以加快检索速度.     

Classify vehicles in traffic scene images with deformable part-based models

Vehicle classification is an important and challenging task in intelligent transportation systems, which has a wide range of applications. In this paper, we propose to integrate vehicle detection and vehicle classification into one single framework by using deformable part-based models. First of all, we use annotated vehicle images to train a deformable part-based model for each class of vehicles to be classified. Then, given a traffic scene image, we employ the obtained vehicle models to perform vehicle detection in it for vehicle extraction. After that, model alignment is performed on the extracted image crop, based on which features are extracted for creating a representation for the vehicle in the given image. We train a linear multi-class Support Vector Machine classifier based on representations of images in a validation set. Finally, we adopt the SVM classifier for vehicle classification. The proposed method is evaluated on the BIT-Vehicle Dataset, and can achieve an accuracy of \(91.08\%\), which is superior to methods used for comparison. Obtained results demonstrated the effectiveness of the proposed method.     

The recovery of three-dimensional structure from image curves

A class of inferences is described which allows the recovery of three-dimensional structures from the two-dimensional curves in an image. Unlike most previous methods, these inferences do not require restrictive assumptions or prior knowledge regarding the scene. They are based on the assumption that the camera viewpoint and the positions of the illumination sources are independent of the objects in the scene. From these independence assumptions, it can be shown that many potential interpretations of image curves are highly improbable. By eliminating these improbable interpretations it is possible to segment the image into sets of related image features and derive many three-space relations.     

Invariant-Based Recognition of Complex Curved 3D Objects from Image Contours

This paper addresses the problem of recognizing three-dimensional objects bounded by smooth curved surfaces from image contours found in a single photograph. The proposed approach is based on a viewpoint-invariant relationship between object geometry and certain image features under weak perspective projection. The image features themselves are viewpoint-dependent. Concretely, the set of all possible silhouette bitangents, along with the contour points sharing the same tangent direction, is the projection of a one-dimensional set of surface points where each point lies on the occluding contour for a five-parameter family of viewpoints. These image features form a one-parameter family of equivalence classes, and it is shown that each class can be characterized by a set of numerical attributes that remain constant across the corresponding five-dimensional set of viewpoints. This is the basis for describing objects by “invariant” curves embedded in high-dimensional spaces. Modeling is achieved by moving an object in front of a camera and does not require knowing the object-to-camera transformation; nor does it involve implicit or explicit three-dimensional shape reconstruction. At recognition time, attributes computed from a single image are used to index the model database, and both qualitative and quantitative verification procedures eliminate potential false matches. The approach has been implemented and examples are presented.     

结合模型集成与特征融合的图像拷贝检测

基于内容的图像拷贝检测关键在于提取的图像特征能够针对不同形式的图像拷贝攻击具有不变性。现实中拷贝攻击手段变化多样,且存在很多相似图像的干扰,目前并没有任何一种图像特征可以对抗所有不同形式的图像攻击。现有方法虽然在图像特征表示上做了很多改进,但都局限于单个特征表示。因此从特征融合的角度对提取特征进行增强,基于卷积神经网络融合图像高层特征以及低层特征以实现特征多样性,集成ImageNet预训练分类模型以及提出的距离度量模型以实现特征互补性。度量模型针对该类问题在预训练模型的基础上通过学习合适的距离度量来对抗由于图像编辑引起的特征差异,拉近拷贝图像与原始图像在特征空间的距离。实验结果表明,结合模型集成和多层深度特征融合的方式可以有效增强特征的鲁棒性,相比单一特征的检测效果提升十分明显。     

基于万有引力定律的人脸识别方法

把万有引力定律引入到人脸识别中,构建基于万有引力定律的人脸识别模型。用传统的主成分分析方法对图像进行特征提取,建立图像与图像之间的相似度模型以及各个类别对图像潜在作用力的表达式,根据该作用力大小决定图像所属类别。在ORL人脸数据库上的实验表明,该模型有一定的可行性。     

Semantic association enhancement transformer with relative position for image captioning

Transformer-based architectures have shown encouraging results in image captioning. They usually utilize self-attention based methods to establish the semantic association between objects in an image for predicting caption. However, when appearance features between the candidate object and query object show weak dependence, the self-attention based methods are hard to capture the semantic association between them. In this paper, a Semantic Association Enhancement Transformer model is proposed to address the above challenge. First, an Appearance-Geometry Multi-Head Attention is introduced to model a visual relationship by integrating the geometry features and appearance features of the objects. The visual relationship characterizes the semantic association and relative position among the objects. Secondly, a Visual Relationship Improving module is presented to weigh the importance of appearance feature and geometry feature of query object to the modeled visual relationship. Then, the visual relationship among different objects is adaptively improved according to the constructed importance, especially the objects with weak dependence on appearance features, thereby enhancing their semantic association. Extensive experiments on MS COCO dataset demonstrate that the proposed method outperforms the state-of-the-art methods.

     

迁移学习在医学图像分类中的研究进展

医学影像作为医疗数据的主要载体,在疾病预防、诊断和治疗中发挥着重要作用。医学图像分类是医学影像分析的重要组成部分。如何提高医学图像分类效率是一个持续的研究问题。随着计算机技术进步,医学图像分类方法已经从传统方法转到深度学习,再到目前热门的迁移学习。虽然迁移学习在医学图像分类中得到较广泛应用,但存在不少问题,本文对该领域的迁移学习应用情况进行综述,从中总结经验和发现问题,为未来研究提供线索。1）对基于迁移学习的医学图像分类研究的重要文献进行梳理、分析和总结,概括出3种迁移学习策略,即迁移模型的结构调整策略、参数调整策略和从迁移模型中提取特征的策略;2）从各文献研究设计的迁移学习过程中提炼共性,总结为5种迁移学习模式,即深度卷积神经网络（deep convolution neural network,DCNN）模式、混合模式、特征组合分类模式、多分类器融合模式和二次迁移模式。阐述了迁移学习策略和迁移学习模式之间的关系。这些迁移学习策略和模式有助于从更高的抽象层次展现迁移学习应用于医学图像分类领域的情况;3）阐述这些迁移学习策略和模式在医学图像分类中的具体应用,分析这些策略及模式的优点、局限性及适用场景;4）给出迁移学习在医学图像分类应用中存在的问题并展望未来研究方向。     

Recovering Relative Depth from Low-Level Features Without Explicit T-junction Detection and Interpretation

This work presents a novel computational model for relative depth order estimation from a single image based on low-level local features that encode perceptual depth cues such as convexity/concavity, inclusion, and T-junctions in a quantitative manner, considering information at different scales. These multi-scale features are based on a measure of how likely is a pixel to belong simultaneously to different objects (interpreted as connected components of level sets) and, hence, to be occluded in some of them, providing a hint on the local depth order relationships. They are directly computed on the discrete image data in an efficient manner, without requiring the detection and interpretation of edges or junctions. Its behavior is clarified and illustrated for some simple images. Then the recovery of the relative depth order on the image is achieved by global integration of these local features applying a non-linear diffusion filtering of bilateral type. The validity of the proposed features and the integration approach is demonstrated by experiments on real images and comparison with state-of-the-art monocular depth estimation techniques.     

自上而下注意图分割的细粒度图像分类

目的 针对细粒度图像分类中的背景干扰问题,提出一种利用自上而下注意图分割的分类模型。方法 首先,利用卷积神经网络对细粒度图像库进行初分类,得到基本网络模型。再对网络模型进行可视化分析,发现仅有部分图像区域对目标类别有贡献,利用学习好的基本网络计算图像像素对相关类别的空间支持度,生成自上而下注意图,检测图像中的关键区域。再用注意图初始化GraphCut算法,分割出关键的目标区域,从而提高图像的判别性。最后,对分割图像提取CNN特征实现细粒度分类。结果 该模型仅使用图像的类别标注信息,在公开的细粒度图像库Cars196和Aircrafts100上进行实验验证,最后得到的平均分类正确率分别为86.74%和84.70%。这一结果表明,在GoogLeNet模型基础上引入注意信息能够进一步提高细粒度图像分类的正确率。结论 基于自上而下注意图的语义分割策略,提高了细粒度图像的分类性能。由于不需要目标窗口和部位的标注信息,所以该模型具有通用性和鲁棒性,适用于显著性目标检测、前景分割和细粒度图像分类应用。     

深度置信网络在极化SAR图像分类中的应用

目的 深度置信网络能够从数据中自动学习、提取特征,在特征学习方面具有突出优势。极化SAR图像分类中存在海量特征利用率低、特征选取主观性强的问题。为了解决这一问题,提出一种基于深度置信网络的极化SAR图像分类方法。方法 首先进行海量分类特征提取,获得极化类、辐射类、空间类和子孔径类四类特征构成的特征集;然后在特征集基础上选取样本并构建特征矢量,用以输入到深度置信网络模型之中;最后利用深度置信网络的方法对海量分类特征进行逐层学习抽象,获得有效的分类特征进行分类。结果 采用AIRSAR数据进行实验,分类结果精度达到91.06%。通过与经典Wishart监督分类、逻辑回归分类方法对比,表现了深度置信网络方法在特征学习方面的突出优势,验证了方法的适用性。结论 针对极化SAR图像海量特征的选取与利用,提出了一种新的分类方法,为极化SAR图像分类提供了一种新思路,为深度置信网络获得更广泛地应用进行有益的探索和尝试。     

多元纹理图像构造方法及应用

多元图像分析方法在图像展开阶段会丢失像素间空间关联关系,导致其对图像纹理特征的分析能力不足.针对此问题,本文提出了一种基于图像纹理特征的多元图像构造方法,并应用于图像分割.首先,结合滑动窗口法和灰度共生矩阵求取图像各通道的纹理特征影像,叠加纹理特征影像构造多元图像.然后,应用多元图像分析方法对所得多元图像进行分析,分割出感兴趣区域.最后,利用分割结果构造决策树模型,以完成对同类感兴趣区域的分割.在图像数据集上进行仿真实验,实验结果表明,本文所提方法的均交并比(MIoU)与同类方法相比有10%左右的提升.