基于 GaborSIFT+NNScSPM 图像特征抽取算法研究

视觉信息的特征表示是计算机视觉场景图像理解研究中的核心内容. 基于GaborSIFT+NNScSPM的图像特征抽取算法,借鉴生物视觉机制中的相关 研究成果,有机结合了HMAX层次计算模型的思想和非负稀疏编码的策略, 较为合理地模拟了生物视觉皮层中视觉处理的过程.在15类场景图 像和Caltech101两个公开数据集上进行了实验验证, 实验结果表明我们所提出的算法较同期算法有着良好的分类性能.     

基于局部特征的非负稀疏编码神经网络模型

在非负稀疏编码(NNSC)的基础上,考虑特征基向量的稀疏度约束和特征基的局部性,提出一种基于局部特征的NNSC神经网络模型。该模型利用梯度和倍增因子相结合的优化算法实现特征系数的学习;利用倍增算法实现特征基的学习。对掌纹图像进行特征提取测试,结果表明,与传统NNSC模型和局部非负矩阵分解(LNMF)方法相比,该模型能有效提取图像的局部特征,收敛速度较快,可模拟初级视觉系统处理自然界信息的稀疏编码策略。     

基于纹理图像分析的生物视觉模型不变性评价

基于纹理图像,从计算机视觉角度对生物视觉模型——视皮层目标识别的标准模型进行定量分析与评价。对原始图像分别进行尺度、旋转及仿射等变化,利用标准模型提取变化后图像的生物视觉特征,再根据提取的生物视觉特征对纹理图像进行分类,采用图像分类结果的曲线下面积来定量分析和评价生物视觉模型是否具有不变性。大量与局部二元模式特征的对比实验表明,该模型提取的生物视觉特征对于纹理图像具备优良的尺度、旋转与仿射不变性。     

结构相似度稀疏编码及其图像特征提取

将结构相似度引入到稀疏编码模型中,提出基于结构相似度的稀疏编码模型。基于该模型提取出图像的稀疏编码特征。实验结果表明,改进后的稀疏编码模型更好地保持了结构信息,更加符合人眼视觉系统特性。将文中提出的模型应用到特征提取中,可获得结构信息保持得更好的图像特征。     

基于稀疏学习与显著性的人脸识别方法研究

针对复杂环境下人脸识别效果差、识别率低等问题,提出一种利用稀疏学习及显著性理论提取人脸特征的方法。基于稀疏编码理论模拟人类视觉感知机制,利用学习得到的基函数构造多尺度多方向滤波器提取图像外观轮廓特征,并对该特征做局部二值模式滤波,以突出人脸局部细节纹理特征。依据视觉注意机制对处理后的特征构造显著图,增强关键特征对于人脸识别的贡献。在LFW、YALE标准库和自制视频帧图像库上的实验结果表明,该方法的识别率高于传统特征提取方法,得到的人脸特征更具代表性,并且在复杂环境下具有较强的鲁棒性。     

基于结构相似度的稀疏编码模型

已有的稀疏编码模型采用误差的平方和作为信息保持的客观评价标准,但最近的研究表明,人眼视觉系统的主要功能是从视觉区域提取图像和视频中的结构化信息.引入结构相似度来衡量信息保持的程度,通过对改进的目标函数进行优化,获得与初级视皮层中具有局部性、朝向性和带通性的感受野相类似的基函数集.实验结果表明,改进后的稀疏编码模型更符合人眼视觉系统特性.     

基于视觉图像的舞姿自动空间分割系统设计

为了提高舞姿的视觉指导效果,需要进行舞姿的自动空间规划设计,提出一种基于视觉图像的舞姿自动空间分割方法。构建舞姿自动空间规划的视觉图像采集模型,采用高分辨的视觉信息增强技术进行舞姿自动空间视觉图像的信息增强处理,提取舞姿自动空间视觉特征量,采用三维视觉重构技术进行舞姿自动空间规划设计,构建舞姿自动空间规划的视觉图像多层网格区域分布模型,结合联合稀疏结构特征分解方法,进行舞姿视觉图像的融合和自适应分割处理,根据特征分割结果进行舞姿的多层空间结构规划和融合聚类,实现舞姿的自动空间分割的算法优化设计。在多层B/S构架体系下实现舞姿自动空间分割系统的软件开发设计。仿真结果表明,采用该方法能有效实现视觉图像下的舞姿自动空间分割,输出峰值信噪比较高,具有较强的舞姿指导性。     

基于生物视觉机理的数字文献图像去噪

针对数字文献图像去噪问题,提出一种基于生物视觉机理的图像去噪算法。模拟初级视皮层简单细胞感受野的响应特性,通过提取数字文献图像的特征,获得数字文献图像的基函数。由动物视觉感知系统的稀疏性,计算神经元对含噪声图像的响应,结合稀疏编码收缩法对响应系数进行收缩,通过响应强烈的神经元重构图像。实验结果表明,与传统的去噪方法相比,该算法能更好地去除数字图像中的高斯噪声,并保留图像细节信息。     

基于多视觉词典的显著性加权图像检索方法

针对视觉词典在图像表示与检索方面的应用需求,本文提出了一种基于多视觉词典与显著性加权相结合的图像检索方法,实现了图像多特征的显著性稀疏表示。该方法首先划分图像为小块,提取图像块的多种底层特征,然后将其作为输入向量,通过非负稀疏编码分别学习图像块多种特征对应的视觉词典,将得到的图 像块稀疏向量经过显著性汇总方法引入空间信息并作显著性加权处理,形成整幅图像的稀疏表示,最后采用提出的SDD距离计算方式进行图像检索。在Corel和Caltech通用图像集上进行仿真实验,与单一视觉词典的方法对比,结果表明本文方法能够有效提高图像检索的准确率。     

基于MFF-GAN的图像集视觉总结

现有图像集视觉总结方法主要使用浅层视觉特征,或者直接应用已训练的卷积神经网络模型提取图像深层特征,选取的图像不具代表性。为此,分析并研究图像集视觉总结的图像特征表示方法,提出多特征图融合生成对抗网络(MFF-GAN)模型。该模型中的判别器通过多特征图融合的方式提取图像特征,使提取的特征能表示图像细节和高层语义信息,并在多特征图融合层后添加自编码网络对特征进行降维,避免特征维度灾难问题。NUS-WIDE数据集上的实验结果验证了M FF-GAN模型的有效性,并表明其能有效提升图像集视觉总结多样性。     

基于深度学习中间层卷积特征的图像标注

针对基于深度特征的图像标注模型训练复杂、时空开销大的不足,提出一种由深 度学习中间层特征表示图像视觉特征、由正例样本均值向量表示语义概念的图像标注方法。首 先,通过预训练深度学习模型的中间层直接输出卷积结果作为低层视觉特征,并采用稀疏编码 方式表示图像;然后,采用正例均值向量法为每个文本词汇构造视觉特征向量,从而构造出文 本词汇的视觉特征向量库;最后,计算测试图像与所有文本词汇的视觉特征向量相似度,并取 相似度最大的若干词汇作为标注词。多个数据集上的实验证明了所提出方法的有效性,就 F1 值而言,该方法在 IAPR TC-12 数据集上的标注性能比采用端到端深度特征的 2PKNN 和 JEC 分 别提高 32%和 60%。     

面向视频语义分析的局部敏感的可鉴别稀疏表示

视频语义分析已经成为人们研究的热点。在传统稀疏表示方法中,相似视频特征未必能产生相近稀疏表示结果。在基于稀疏表示的视频语义分析中,假定相似的视频数据样本的稀疏表示也相似,即两个相似视频特征的稀疏系数之间的距离较小。为了提高视频语义分析的准确性,基于该假设提出一种面向视频语义分析的局部敏感的可鉴别稀疏表示方法。该方法在局部敏感稀疏表示中引入基于稀疏系数的鉴别损失函数,优化构建稀疏表示的字典,使稀疏表示特征满足类内离散度小、类间离散度大的Fisher准则,并建立可鉴别稀疏模型。为验证所提方法的有效性,在相关视频数据库中将其与多种算法进行对比,实验结果表明,该方法显著地提高了视频特征稀疏表示的鉴别性,有效地提高了视频语义分析的准确性。     

基于多层非负局部Laplacian稀疏编码的图像分类

针对单层稀疏编码结构对图像特征学习能力的局限性问题,提出了一个基于图像块稀疏表示的深层架构,即多层融合局部性和非负性的Laplacian稀疏编码算法（MLLSC）。对每个图像平均区域划分并进行尺度不变特征变换（SIFT）特征提取,在稀疏编码阶段,在Laplacian稀疏编码的优化函数中添加局部性和非负性,在第一层和第二层分别进行字典学习和稀疏编码,分别得到图像块级、图像级的稀疏表示,为了去除冗余特征,在进行第二层稀疏编码之前进行主成分分析（PCA）降维,最后采用多类线性支持向量机进行分类。在四个标准数据集上进行验证,实验结果表明,MLLSC方法具有高效的特征学习能力,能够捕获图像更深层次的特征信息,相对于单层结构算法准确率提高了3%~13%,相对于多层稀疏编码算法准确率提高了1%~2.3%;并对不同参数进行了对比分析,充分展现了其在图像分类中的有效性。     

Finding independent components using spikes: A natural result of hebbian learning in a sparse spike coding scheme

As an alternative to classical representations in machine learning algorithms, we explore coding strategies using events as is observed for spiking neurons in the central nervous system. Focusing on visual processing, we have previously shown that we can define with an over-complete dictionary a sparse spike coding scheme by implementing lateral interactions that account for redundant information. Since this class of algorithms is both compatible with biological constraints and with neuro-physiological observations, it can provide a possible algorithm to explain the speed of visual processing despite the relatively slow time of response of single neurons. Here, I explore learning mechanisms to derive in an unsupervised manner an over-complete set of filters which provides a progressively sparser representation of the input. This work is based on a previous model of sparse coding from Olshausen et al. (1998) and the results leads to similar results, suggesting that this strategy provides a simple neural implementation of this algorithm and thus of Blind Source Separation. Moreover, this neuro-mimetic algorithm may be easily extended to realistic architectures of cortical columns in the primary visual cortex and we show results for different strategies of representation, leading to neuro-mimetic adaptive sparse spike coding schemes. This revised version was published online in June 2006 with corrections to the Cover Date.     

Image classification using label constrained sparse coding

Sparse coding has been widely used for feature encoding in recent years. However, the encoded parameters’ similarity is ignored with sparse coding. Besides, the label information from which class the local feature is extracted is also ignored. To solve this problem, in this paper, we propose a novel feature encoding method called label constrained sparse coding (LCSC) for visual representation. The visual similarities between local features are jointly considered with the corresponding label information of local features. This is achieved by combining the label constraints with the encoding of local features. In this way, we can ensure that similar local features with the same label are encoded with similar parameters. Local features with different labels are encoded with dissimilar parameters to increase the discriminative power of encoded parameters. Besides, instead of optimizing for the coding parameter of each local feature separately, we jointly encode the local features within one sub-region in the spatial pyramid way to combine the spatial and contextual information of local features. We apply this label constrained sparse coding technique for classification tasks on several public image datasets to evaluate its effectiveness. The experimental results shows the effectiveness of the proposed method.     

改进的非负稀疏编码图像基学习算法

图像基学习是图像特征提取与表示的重要方法之一。非负稀疏编码不仅具有标准稀疏编码算法的自适应性、空间的局部性、方向性和频域的带通性,而且更能反应哺乳动物的视觉机制。本文在非负稀疏编码的基础上,利用经验模态分解技术加入了图像的结构信息,提出了结合经验模态分解的非负稀疏编码算法,保证了系数矩阵的稀疏性与所提取图像特征的结构性。学习得到的图像基不仅具有非负稀疏编码的特征,而且更好地表示出图像的结构信息。     

The image annotation algorithm using convolutional features from intermediate layer of deep learning

The automatic image annotation is an effective computer operation that predicts the annotation of an unknown image by automatically learning potential relationships between the semantic concept space and the visual feature space in the annotation image dataset. Usually, the auto-labeling image includes the processing: learning processing and labeling processing. Existing image annotation methods that employ convolutional features of deep learning methods have a number of limitations, including complex training and high space/time expenses associated with the image annotation procedure. Accordingly, this paper proposes an innovative method in which the visual features of the image are presented by the intermediate layer features of deep learning, while semantic concepts are represented by mean vectors of positive samples. Firstly, the convolutional result is directly output in the form of low-level visual features through the mid-level of the pre-trained deep learning model, with the image being represented by sparse coding. Secondly, the positive mean vector method is used to construct visual feature vectors for each text vocabulary item, so that a visual feature vector database is created. Finally, the visual feature vector similarity between the testing image and all text vocabulary is calculated, and the vocabulary with the largest similarity used for annotation. Experiments on the datasets demonstrate the effectiveness of the proposed method; in terms of F1 score, the proposed method’s performance on the Corel5k dataset and IAPR TC-12 dataset is superior to that of MBRM, JEC-AF, JEC-DF, and 2PKNN with end-to-end deep features.

     

Non-negativity and locality constrained Laplacian sparse coding for image classification

The traditional sparse coding (SC) method has achieved good results in image classification. However, one of its serious weaknesses is that it ignores the relationship between features thus losing spatial information. Moreover, in combinatorial optimisation problems, operations of addition and subtraction are involved, and the use of subtraction may cause features to be cancelled. In this paper, we propose a method called non-negativity and locality constrained Laplacian sparse coding (NLLSC) for image classification. Firstly, non-negative matrix factorisation (NMF) is used in the Laplacian sparse coding (LSC), which is applied to constrain the negativity of both codebook and code coefficient. Secondly, we introduce K-nearest neighbouring codewords for local features because locality is more important than sparseness. Finally, non-negativity and locality constrained operators are introduced to obtain a novel sparse coding for local features, and then in the pooling step, we use spatial pyramid division (SPD) and max pooling (MP) to represent the final images. As for image classification, multi-class linear SVM is adopted. Experiments on several standard image datasets have shown better performance than previous algorithms.