基于MKSVM的多示例学习算法及刑侦图像分类

针对刑侦图像分类问题,提出一种基于多核支持向量机的多示例学习(MIL)算法.首先,该方法采用金字塔网格划分法对刑侦图像进行分块,再将每幅图像作为一个多示例包,每个子块的底层视觉特征作为包中的示例,将刑侦图像分类问题转化为MIL问题;然后,采用K-means双重聚类方法对所有多示例包进行聚类生成聚类中心并定义为视觉字,再把视觉字的集合构造成视觉投影空间;最后,通过设计的非线性投影函数将每个包映射为视觉投影空间中的一个点,则MIL问题被转化为一个标准的有监督学习问题,并采用多核支持向量机(MKSVM)来训练刑侦图像分类器.基于真实刑侦图像库的对比实验表明,所提方法具有较好的鲁棒性,且分类精度高于其他方法.     

LSA based multi-instance learning algorithm for image retrieval

Focusing on the problem of natural image retrieval, based on latent semantic analysis (LSA) and support vector machine (SVM), a novel multi-instance learning (MIL) algorithm is proposed, where a bag corresponds to an image and an instance corresponds to the low-level visual features of a segmented region. Firstly, in order to transform every bag into a single sample, a collection of “visual-word” is generated by k-means clustering method to construct a projection space, then a nonlinear mapping is defined using these “visual-word” to embed each bag as a point in the projection space, thereby obtaining every bag's projection feature. Secondly, the matrix consisted of all the projection features of training bags is regarded as a term-document matrix, and LSA method is used to obtain the latent semantic feature of each bag. As a result, the MIL problem is converted into a standard single instance learning (SIL) problem that can be solved directly by SVM method. Experimental results on the COREL data sets show that the proposed method, named LSASVM-MIL, is robust, and its performance is superior to other key existing MIL algorithms.     

基于EMD-CkNN多示例学习算法的图像分类

针对自然图像场景分类问题,根据Citation-kNN算法思想,提出一种新的基于多示例学习(MIL)的图像分类方法。将整个图像当作多示例包,图像分割的区域当作包中的示例,在度量图像包间的相似距离时,利用改进的推土机距离(EMD)代替Citation-KNN算法中的最小Hausdorff距离(minHD),用于图像分类。在Corel图像库上的对比实验结果表明,分类准确率更高。     

基于稀疏编码和集成学习的多示例多标记图像分类方法

 该文基于稀疏编码和集成学习提出了一种新的多示例多标记图像分类方法。首先,利用训练包中所有示例学习一个字典,根据该字典计算示例的稀疏编码系数;然后基于每个包中所有示例的稀疏编码系数计算包特征向量,从而将多示例多标记问题转化为多标记问题;最后利用多标记分类算法进行求解。为了提高分类器的泛化能力,对多个分类器进行集成。在多示例多标记图像数据集上的实验结果表明所提方法与其它方法相比有更好的性能。     

Seeing is believing: video classification for computed tomographic colonography using multiple-instance learning

In this paper, we present development and testing results for a novel colonic polyp classification method for use as part of a computed tomographic colonography (CTC) computer-aided detection (CAD) system. Inspired by the interpretative methodology of radiologists using 3-D fly-through mode in CTC reading, we have developed an algorithm which utilizes sequences of images (referred to here as videos) for classification of CAD marks. For each CAD mark, we created a video composed of a series of intraluminal, volume-rendered images visualizing the detection from multiple viewpoints. We then framed the video classification question as a multiple-instance learning (MIL) problem. Since a positive (negative) bag may contain negative (positive) instances, which in our case depends on the viewing angles and camera distance to the target, we developed a novel MIL paradigm to accommodate this class of problems. We solved the new MIL problem by maximizing a L2-norm soft margin using semidefinite programming, which can optimize relevant parameters automatically. We tested our method by analyzing a CTC data set obtained from 50 patients from three medical centers. Our proposed method showed significantly better performance compared with several traditional MIL methods.     

Self-learning-based post-processing for image/video deblocking via sparse representation

Blocking artifact, characterized by visually noticeable changes in pixel values along block boundaries, is a common problem in block-based image/video compression, especially at low bitrate coding. Various post-processing techniques have been proposed to reduce blocking artifacts, but they usually introduce excessive blurring or ringing effects. This paper proposes a self-learning-based post-processing framework for image/video deblocking by properly formulating deblocking as an MCA (morphological component analysis)-based image decomposition problem via sparse representation. Without the need of any prior knowledge (e.g., the positions where blocking artifacts occur, the algorithm used for compression, or the characteristics of image to be processed) about the blocking artifacts to be removed, the proposed framework can automatically learn two dictionaries for decomposing an input decoded image into its “blocking component” and “non-blocking component.” More specifically, the proposed method first decomposes a frame into the low-frequency and high-frequency parts by applying BM3D (block-matching and 3D filtering) algorithm. The high-frequency part is then decomposed into a blocking component and a non-blocking component by performing dictionary learning and sparse coding based on MCA. As a result, the blocking component can be removed from the image/video frame successfully while preserving most original visual details. Experimental results demonstrate the efficacy of the proposed algorithm.     

基于多示例学习的图像检索方法

多示例学习对处理各类歧义问题有较好的效果,将它应用于周像检索问题,提出了一种新的基于多示例学习的图像检索方法。首先提取每幅图像的局部区域特征,通过对这些特征聚类求得一组基向量,并利用它们对每个局部特征向量进行编码,接着使用均值漂移聚类算法对图像进行分割,根据局部特征点位置所对应的分割块划分特征编码到相应的子集,最后将每组编码子集聚合成一个向量,这样每幅图像对应一个多示例包。根据用户选择的图像生成正包和反包,采用多示例学习算法进行学习,取得了较为满意的结果。     

Multiple-instance learning algorithms for computer-aided detection

Many computer-aided diagnosis (CAD) problems can be best modelled as a multiple-instance learning (MIL) problem with unbalanced data, i.e., the training data typically consists of a few positive bags, and a very large number of negative instances. Existing MIL algorithms are much too computationally expensive for these datasets. We describe CH, a framework for learning a Convex Hull representation of multiple instances that is significantly faster than existing MIL algorithms. Our CH framework applies to any standard hyperplane-based learning algorithm, and for some algorithms, is guaranteed to find the global optimal solution. Experimental studies on two different CAD applications further demonstrate that the proposed algorithm significantly improves diagnostic accuracy when compared to both MIL and traditional classifiers. Although not designed for standard MIL problems (which have both positive and negative bags and relatively balanced datasets), comparisons against other MIL methods on benchmark problems also indicate that the proposed method is competitive with the state-of-the-art.     

Water leakage image recognition of shield tunnel via learning deep feature representation

With the development of urban metro, the research on structural diseases of shield tunnels has been becoming a hot research topic, especially the leakage water diseases. Deep learning-based algorithms have shown impressive performance in image processing domain, such as image classification, image recognition or image retrieval. In this paper, we propose a novel image recognition algorithm for water leakage diseases of shield tunnels based on deep learning algorithm. Water leakage images are classified into six categories, each of which are extracted deep representation for image recognition. We compare our method with Otsu algorithm (OA), Region Growing Algorithm (RGA), and Watershed Algorithm (WA) to show the effectiveness of our proposed method.     

一种新的多尺度深度学习图像语义理解方法研究

如何在深度学习中融合 图像的多尺度信息,是基于深度学习的视觉算法需要解决的一个关键问题。本文提出一种基 于多尺度交替 迭代训练的深度学习方法,并应用于图像的语义理解。算法采用卷积神经网络(CNN)从原始 图像中提取稠密性特征 来编码以每个像素为中心的矩形区域,将多个尺度图像交替迭代训练,能够捕获不同尺度下 的纹理、颜色和 边缘等重要信息。在深度学习提取特征分类结果的基础上,提出了一种结合超像素分割的方 法,统计超像 素块的主导类别,来校正分类错误的像素类别,同时描绘出目标区域边界轮廓,完成最终的 语义理解。在Stanford Background Dataset 8类数据集上验证了本文方法的有效性,准确 率达到77.4%。     

Direction-of-Arrival Method Based on Randomize-Then-Optimize Approach

The direction-of-arrival(DOA) estimation problem can be solved by the methods based on sparse Bayesian learning(SBL). To assure the accuracy, SBL needs massive amounts of snapshots which may lead to a huge computational workload. In order to reduce the snapshot number and computational complexity, a randomizethen-optimize(RTO) algorithm based DOA estimation method is proposed. The “learning” process for updating hyperparameters in SBL can be avoided by using the optimization and Metropolis-Hasti...     

Iterative transductive learning for automatic image segmentation and matting with RGB-D data

In this paper, we propose a fully automatic image segmentation and matting approach with RGB-Depth (RGB-D) data based on iterative transductive learning. The algorithm consists of two key elements: robust hard segmentation for trimap generation, and iterative transductive learning based image matting. The hard segmentation step is formulated as a Maximum A Posterior (MAP) estimation problem, where we iteratively perform depth refinement and bi-layer classification to achieve optimal results. For image matting, we propose a transductive learning algorithm that iteratively adjusts the weights between the objective function and the constraints, overcoming common issues such as over-smoothness in existing methods. In addition, we present a new way to form the Laplacian matrix in transductive learning by ranking similarities of neighboring pixels, which is essential to efficient and accurate matting. Extensive experimental results are reported to demonstrate the state-of-the-art performance of our method both subjectively and quantitatively.     

一种基于稀疏编码的多核学习图像分类方法

 本文提出一种基于稀疏编码的多核学习图像分类方法.传统稀疏编码方法对图像进行分类时,损失了空间信息,本文采用对图像进行空间金字塔多划分方式为特征加入空间信息限制.在利用非线性SVM方法进行图像分类时,空间金字塔的各层分别形成一个核矩阵,本文使用多核学习方法求解各个核矩阵的权重,通过核矩阵的线性组合来获取能够对整个分类集区分能力最强的核矩阵.实验结果表明了本文所提出图像分类方法的有效性和鲁棒性.对Scene Categories场景数据集可以达到83.10%的分类准确率,这是当前该数据集上能达到的最高分类准确率.     

基于迁移集成学习的无人机图像识别算法

针对当前小型无人机目标图像识别方法准确率较低的问题,提出了一种基于迁移集成学习的无人机图像识别算法。首先,基于AlexNet、VGGNet-19、Inception-V3以及ResNet-50四种结构具有差异的卷积神经网络对源数据集进行预训练,获取图像的深层次特征;然后,对目标数据集进行迁移学习,得到目标的分类特征,构建分类模型;之后,采用相对多数投票法和加权平均法的集成学习方法,对分类模型进行集成得到迁移集成模型。构建了一个包含小型无人机图像、飞鸟图像以及直升机图像的图像数据集UavNet,在对数据集进行数据增强的基础上开展了图像识别算法性能实验,结果表明,算法对多类目标的识别准确率为99.42%,无人机类目标识别的F1-score指标为99.12%,优于主流的卷积神经网络方法和传统的支持向量机方法,具有一定的理论意义和应用价值。     

利用0-1矩阵分解集成的极化SAR图像分类

全极化合成孔径雷达(PolSAR)图像蕴含更丰富的散射信息,具有更多的可用特征。如何使用这些特征是极化SAR图像分类中非常重要的一步,但是目前尚未对此提出非常明确的准则。为了能够有效地解决上述问题,该文提出一种基于特征加权集成的极化SAR图像分类算法。该算法采用0-1矩阵分解集成方法对包括不同特征的数据集进行学习获得相应加权系数,并通过对每个特征集获得的预测结果进行加权集成来提高极化SAR图像分类性能。首先,输入极化SAR数据,获得极化特征作为原始特征集,并对其进行随机抽取获得不同的特征子集;然后,使用0-1矩阵集成算法得到每个特征值相对应的加权系数;最后,通过对各个特征子集的预测结果进行集成得到最终极化SAR图像分类结果。实测L波段和C波段极化数据的实验结果表明,该算法可以有效地提高极化SAR图像分类的准确度。     

Spatial-invariant convolutional neural network for photographic composition prediction and automatic correction

“Composition” determines the vividness of the image and its narrative power. Current research on image aesthetics implicitly considers simple composition rules, but no reliable composition classification and image optimization method explicitly considers composition rules. The existing composition classification models are not suitable for snapshots. We propose a composition classification model based on spatial-invariant convolutional neural networks (RSTN) with translation invariance and rotation invariance. It enhances the generalization of the model for snapshots or skewed images. Ultimately, the accuracy of the RSTN model improved by 3% over the Baseline to 90.8762%, and the rotation consistency improved by 16.015%. Furthermore, we classify images into three categories based on their sensitivity to editing: skew-sensitive, translation-sensitive, and non-space-sensitive. We design a set of composition optimization strategies for each composition that can effectively adjust the composition to beautify the image.     

基于粒子群算法寻最优属性关联下的零样本语义自编码器

针对零样本图像分类构建共享属性层时造成的信息缺失问题,该文提出一种嵌入属性关联性的补偿方法.通过语义自编码器构建特征到属性的映射,然后以最大后验概率估计在类高斯模型构建的基础上实现零样本图像分类.为弥补SAE对属性关系学习的不足,引入加性因子与乘性因子对属性相关性进行嵌入,并利用粒子群算法搜寻最优的因子参数,实现属性相关性信息的补偿.实验结果表明采取相同映射方法的情况下,基于属性相关性嵌入的零样本图像分类在Pubfig数据集和OSR数据集上的分类效果较之其他方法得到了显著提升.     

基于“图像”分割的小波脊线提取算法

针对传统脊线提取算法不能同时兼顾速度和精度的问题,本文提出了一种新的基于“图像”分割的小波脊线提取算法。对渐近性信号进行连续小波变换以后,模值较大的小波系数往往集中在时间-尺度平面上几个分散的区域,将小波系数模值矩阵看作一个“图像”,对其分割,再对分割得到的每个区域确定其极值位置可得到小波脊线。仿真实验表明：本文算法不仅较传统脊线算法在精度和效率都有所提高,在信号去噪和信号分离中也表现良好。     

哈希快速多标记学习算法

本文针对多标记学习耗时大、很难处理大规模数据的问题,提出了一种哈希快速多标记学习算法（HFMLL）,该算法将哈希算法与多标记学习算法结合,采用局部敏感哈希算法快速获得每个样本的近邻样本,并通过最小独立置换的MinHash算法快速找到每个标记的相关标记,根据其近邻样本及相关标记的信息,运用最大后验概率准则来预测新样本的标记集。实验表明HFMLL 算法在保持较高分类性能的情况下,算法速度明显优于目前的多标记算法,可以广泛应用于大规模的数据集。      

自适应稀疏约束图像超分辨力重建方法

简要介绍了基于稀疏字典约束的超分辨力重建算法,提出了具有低复杂度的基于K均值聚类的自适应稀疏约束图像超分辨力重建算法。所提算法从两个方面降低其计算复杂度:分类训练字典,对图像块归类重建,降低每个图像块所用字典的大小;对图像块的特征进行分析,自适应地选择重建方法。实验结果表明,提出的快速重建方法在重建质量与原算法相当的前提下,可以较大程度地降低重建时间。