基于多标签学习的卷积神经网络的图像标注方法

针对图像自动标注中因人工选择特征而导致信息缺失的缺点,提出使用卷积神经网络对样本进行自主特征学习。为了适应图像自动标注的多标签学习的特点以及提高对低频词汇的召回率,首先改进卷积神经网络的损失函数,构建一个多标签学习的卷积神经网络（CNN-MLL）模型,然后利用图像标注词间的相关性对网络模型输出结果进行改善。通过在IAPR TC-12标准图像标注数据集上对比了其他传统方法,实验得出,基于采用均方误差函数的卷积神经网络（CNN-MSE）的方法较支持向量机（SVM）方法在平均召回率上提升了12.9%,较反向传播神经网络（BPNN）方法在平均准确率上提升了37.9%;基于标注结果改善的CNN-MLL方法较普通卷积神经网络的平均准确率和平均召回率分别提升了23%和20%。实验结果表明基于标注结果改善的CNN-MLL方法能有效地避免因人工选择特征造成的信息缺失同时增加了对低频词汇的召回率。     

基于形式概念分析的图像场景语义标注模型

为生成有效表示图像场景语义的视觉词典,提高场景语义标注性能,提出一种基于形式概念分析(FCA)的图像场景语义标注模型。该方法首先将训练图像集与其初始的视觉词典抽象为形式背景,采用信息熵标识了各视觉单词的权重,并分别构造了各场景类别概念格结构;然后再利用各视觉单词权重的均值刻画概念格内涵上各组合视觉单词标注图像的贡献,按照类别视觉词典生成阈值,从格结构上有效提取了标注各类场景图像语义的视觉词典;最后,利用K最近邻标注测试图像的场景语义。在Fei-Fei Scene 13类自然场景图像数据集上进行实验,并与Fei-Fei方法和Bai方法相比,结果表明该方法在β=0.05和γ=15时,标注分类精度更优。     

结合相关反馈日志与语义网络的图像标注

连接高层语义和低层视觉特征的图像语义标注技术能够很好地表示图像的语义,提出并实现了一种结合相关反馈日志与语义网络的图像标注方法。该方法以收集的用户相关反馈日志为基础获得图像的语义信息,通过计算图像间的语义相似度进行语义聚类并采用语义传播的方式实现图像的语义标注。实验结果表明,随着相关反馈日志库的不断扩充,图像库中越来越多的图像会在反馈的过程中得到标注且标注的准确率会随着反馈次数的增加而趋于稳定。     

基于生成式对抗网络的图像自动标注

针对基于深度学习的图像标注模型输出层神经元数目与标注词汇量成正比，导致模型结构因词汇量的变化而改变的问题，提出了结合生成式对抗网络（GAN）和Word2vec的新标注模型。首先，通过Word2vec将标注词汇映射为固定的多维词向量；其次，利用GAN构建神经网络模型--GAN-W模型，使输出层神经元数目与多维词向量维数相等，与词汇量不再相关；最后，通过对模型多次输出结果的排序来确定最终标注。GAN-W模型分别在Corel 5K和IAPRTC-12图像标注数据集上进行实验，在Corel 5K数据集上，GAN-W模型准确率、召回率和F1值比卷积神经网络回归（CNN-R）方法分别提高5、14和9个百分点；在IAPRTC-12数据集上，GAN-W模型准确率、召回率和F1值比两场K最邻近（2PKNN）模型分别提高2、6和3个百分点。实验结果表明，GAN-W模型可以解决输出神经元数目随词汇量改变的问题，同时每幅图像标注的标签数目自适应，使得该模型标注结果更加符合实际标注情形。     

Support vector description of clusters for content-based image annotation

Continual progress in the fields of computer vision and machine learning has provided opportunities to develop automatic tools for tagging images; this facilitates searching and retrieving. However, due to the complexity of real-world image systems, effective and efficient image annotation is still a challenging problem. In this paper, we present an annotation technique based on the use of image content and word correlations. Clusters of images with manually tagged words are used as training instances. Images within each cluster are modeled using a kernel method, in which the image vectors are mapped to a higher-dimensional space and the vectors identified as support vectors are used to describe the cluster. To measure the extent of the association between an image and a model described by support vectors, the distance from the image to the model is computed. A closer distance indicates a stronger association. Moreover, word-to-word correlations are also considered in the annotation framework. To tag an image, the system predicts the annotation words by using the distances from the image to the models and the word-to-word correlations in a unified probabilistic framework. Simulated experiments were conducted on three benchmark image data sets. The results demonstrate the performance of the proposed technique, and compare it to the performance of other recently reported techniques.     

An image recognition method by rough classification for a scene image

We have recognized the regions of scene images for image recognition. First, the proposed segmentation method classifies images into several segments without using the Euclidian distance. We need several features to recognize regions. However, they are different for chromatic and achromatic colors. The regions are divided into three categories (black, achromatic, and chromatic). In this article, we focus on the achromatic category. The averages of the intensity and the fractal dimension features of the regions in the achromatic category are calculated. We recognize the achromatic region by using a neural network with suitable features. In order to show the effectiveness of the proposed method, we have recognized the regions. This work was presented in part at the 10th International Symposium on Artificial Life and Robotics, Oita, Japan, February 4–6, 2005     

基于决策融合的图像自动标注方法

结合多媒体描述接口（MPEG-7）和MM（Mixture Model）混合模型,实现了基于决策融合的图像自动标注。在图像标注过程中,分别利用颜色描述子和纹理描述子为每个主题下的图像建立MM混合模型,实现低层视觉特征到高层语义空间的映射,利用局部决策融合方式融合在颜色和纹理MM混合模型下的标注结果,实现图像自动标注。通过在corel图像数据集上的实验,表明提出的局部决策融合方式能更充分利用图像的颜色和纹理信息,提高了图像标注性能。     

New neural network algorithm for image reconstruction from fan-beam projections

Neural networks have some applications in computerized tomography, in particular to reconstruct an image from projections. The presented paper describes a new practical approach to the reconstruction problem using a Hopfield-type neural network. The methodology of this reconstruction algorithm resembles a transformation formula—the so-called ρ-filtered layergram method. The method proposed in this work is adapted for discrete fan beam projections, already used in practice. Performed computer simulations show that the neural network reconstruction algorithm designed to work in this way outperforms conventional methods in obtained image quality, and in perspective of hardware implementation in the speed of the reconstruction process.     

A neural network construction algorithm with application to image compression

We propose an algorithm for constructing a feedforward neural network with a single hidden layer. This algorithm is applied to image compression and it is shown to give satisfactory results. The neural network construction algorithm begins with a simple network topology containing a single unit in the hidden layer. An optimal set of weights for this network is obtained by applying a variant of the quasi-Newton method for unconstrained optimisation. If this set of weights does not give a network with the desired accuracy then one more unit is added to the hidden layer and the network is retrained. This process is repeated until the desired network is obtained. We show that each addition of the hidden unit to the network is guaranteed to increase the signal to noise ratio of the compressed image.     

唐卡图像关键区域对象概念的语义相似度计算

唐卡图像中关键区域对象的概念所表达内涵具有一定的相似性,进行定量计算和分析,对研究唐卡图像高层语义检索具有重要意义。针对该问题,引入形式概念分析,提出一种唐卡图像关键区域概念语义相似度的计算方法。首先提取唐卡图像中关键区域对象的概念和一系列的语义关键词作为形式背景来构造概念格,通过概念格计算概念间的语义相似度,实验结果表明,本文方法计算结果与人工判断结果相吻合,具有可行性和有效性。     

An efficient approach for the removal of impulse noise from the corrupted image using neural network based impulse detector

A new efficient approach to detect the impulse noise from the corrupted image using feed forward neural network (FFNN) is presented. A modified version of the arithmetic mean filter is proposed to remove the detected impulse noise. The performance of proposed noise detection approach is analyzed using the performance measures such as False Alarm Ratio (FAR), Missed Noise (MN) pixels and Falsely Detected Noise (FDN) pixels. The simulation results show that these performances are robust even at higher percentage of noise. The filtered result is compared with the other recent approaches in terms of Peak Signal to Noise Ratio (PSNR). The proposed method produces remarkably good results both in quantitative measures and qualitative judgments of image quality.     

SOM神经网络改进及在遥感图像分类中的应用*

针对自组织特征神经网络自身算法的特点和缺陷,采用遗传算法对网络进行改进,形成了基于遗传算法的自组织特征神经网络,并从输入向量、竞争层神经元数量设置和初始权向量设定三方面,结合遥感图像的特性对自组织特征映射网络遥感图像分类的方法进行了改进。将该方法应用于择西安地区的ETM+卫星遥感图像进行分类试验,结果表明,基于遗传算法的自组织特征映射网络使得遥感图像的分类精度更高,且该算法实现简单,具有一定的工程应用价值。     

应用BP神经网络分类器识别交通标志

介绍了神经网络特性和BP神经网络分类器的一般原理。针对交通标志识别需要处理的信息量大以及受天气道路等外界条件的影响存在噪声干扰的情况，提出了一种应用BP神经网络分类器识别交通标志的方法。识别分为图像数字处理、BP神经网络的训练、测试与对加入噪声图像进行识别3个步骤，经实验取得了良好的识别效果。     

Supervised learning technique for a mobile robot controller in a visual line tracking task

This article deals with the development of learning methods for an intelligent control system for an autonomous mobile robot. On the basis of visual servoing, an approach to learning the skill of tracking colored guidelines is proposed. This approach utilizes a robust and adaptive image processing method to acquire features of the colored guidelines and convert them into the controller input. The supervised learning procedure and the neural network controller are discussed. The method of obtaining the learning data and training the neural network are described. Experimental results are presented at the end of the article. This work was presented, in part, at the Sixth International Symposium on Artificial Life and Robotics, Tokyo, Japan, January 15–17, 2001     

一种网络流量预测的小波神经网络模型

结合小波变换和人工神经网络的优势,建立一种网络流量预测的小波神经网络模型。首先对流量时间序列进行小波分解,得到小波变换尺度系数序列和小波系数序列,以系数序列和原来的流量时间序列分别作为模型的输入和输出,构造人工神经网络并且加以训练。用实际网络流量对该模型进行验证,结果表明,该模型具有较高的预测效果。     

Effects of the number of hidden nodes used in a structured-based neural network on the reliability of image classification

A structured-based neural network (NN) with backpropagation through structure (BPTS) algorithm is conducted for image classification in organizing a large image database, which is a challenging problem under investigation. Many factors can affect the results of image classification. One of the most important factors is the architecture of a NN, which consists of input layer, hidden layer and output layer. In this study, only the numbers of nodes in hidden layer (hidden nodes) of a NN are considered. Other factors are kept unchanged. Two groups of experiments including 2,940 images in each group are used for the analysis. The assessment of the effects for the first group is carried out with features described by image intensities, and, the second group uses features described by wavelet coefficients. Experimental results demonstrate that the effects of the numbers of hidden nodes on the reliability of classification are significant and non-linear. When the number of hidden nodes is 17, the classification rate on training set is up to 95%, and arrives at 90% on the testing set. The results indicate that 17 is an appropriate choice for the number of hidden nodes for the image classification when a structured-based NN with BPTS algorithm is applied.     

混合阶通道注意力网络的单图像超分辨率重建

目前用于图像超分辨率重建的通道注意力机制存在注意力预测破坏每个通道和其权重的直接对应关系以及仅仅只考虑一阶或二阶通道注意力而没有综合考虑优势互补的问题,因此提出一种混合阶通道注意力网络的单图像超分辨率重建算法。首先,该网络框架利用局部跨通道相互作用策略将之前一、二阶通道注意力模型采用的升降维改为核为k的一维卷积。这样不仅使得通道注意力预测更直接准确,而且得到的模型相比之前的通道注意力模型更简单;同时,采用改进一、二阶通道注意力模型以综合利用不同阶通道注意力的优势,提高网络判别能力。在基准数据集上的实验结果表明,和现有的超分辨率算法相比,所提算法重建图像的纹理细节和高频信息能得到更好的恢复,且在Set5和BSD100数据集上感知指数（PI）分别平均提高0.3和0.1。这表明此网络能更准确地预测通道注意力并综合利用了不同阶通道注意力,一定程度上提升了性能。     

条件生成对抗网络的低照度图像增强方法

目的 在日常的图像采集工作中,由于场景光照条件差或设备的补光能力不足,容易产生低照度图像。为了解决低照度图像视觉感受差、信噪比低和使用价值低（难以分辨图像内容）等问题,本文提出一种基于条件生成对抗网络的低照度图像增强方法。方法 本文设计一个具备编解码功能的卷积神经网络（CNN）模型作为生成模型,同时加入具备二分类功能的CNN作为判别模型,组成生成对抗网络。在模型训练的过程中,以真实的亮图像为条件,依靠判别模型监督生成模型以及结合判别模型与生成模型间的相互博弈,使得本文网络模型具备更好的低照度图像增强能力。在本文方法使用过程中,无需人工调节参数,图像输入模型后端到端处理并输出结果。结果 将本文方法与现有方法进行比较,利用本文方法增强的图像在亮度、清晰度以及颜色还原度等方面有了较大的提升。在峰值信噪比、直方图相似度和结构相似性等图像质量评价指标方面,本文方法比其他方法的最优值分别提高了0.7 dB、3.9%和8.2%。在处理时间上,本文方法处理图像的速度远远超过现有的传统方法,可达到实时增强的要求。结论 通过实验比较了本文方法与现有方法对于低照度图像的处理效果,表明本文方法具有更优的处理效果,同时具有更快的处理速度。