人脸图像光照预处理的一种新方法

人脸识别是图像分析和理解领域中最成功的应用之一,近年来得到了迅速的发展,但是阻碍人脸识别技术应用到实际中的瓶颈之一——光照问题,一直没能得到很好的解决。局部二值模式是最近发展起来的一种理论简单但功能强大的纹理分析算法,在计算机视觉等领域表现出良好的性能。将该纹理提取算法应用到图像预处理中并并利用大规模中国人脸图像数据库CAS-PEAL-R1来检验这种方法的有效性。实验结果表明:加入LBP纹理后,该方法能较好解决光照变化问题,提高识别性能。     

An ensemble template algorithm for extracting targets from blurred infrared images

In this paper, an ensemble template algorithm is proposed to extract targets from blurred infrared images. First, the image pixels are divided according to their gray values into three pixel sets, a target set, a background set and the third set without class label. Second, the neighborhood statistical characteristics for each pixel are calculated as its template features. Third, ensemble detectors are designed using target pixels and background pixels based on their template features, and these ensemble detectors are used to detect the third pixel set. To evaluate the performance of the proposed extraction algorithm, this paper compares the ensemble template with other extraction algorithms using blurred infrared image of hand trace. Experimental results show that the ensemble template algorithm proposed in this paper exhibits better extraction performance.     

多分类器融合提取土壤养分特征波长

光谱已经应用于土壤养分速测的分析，但是如何寻找土壤光谱特征波段，尽最大可能避免无用信息干扰、保留有用信息，建立准确度高、预测效果好的模型仍是一个亟需解决的问题。以青岛三个不同地区土壤样品为例，测定土壤样品的紫外-可见-近红外光谱及其总碳（TC）、总氮（TN）、总磷（TP）含量；分别采用连续投影算法（SPA）、无信息变量消除法（UVE）、遗传算法（GA）、相关系数法（CC）四种算法（四种单分类器）对土壤光谱提取特征波长；再引入投票法和加权投票法的多分类器融合方法将四种算法融合得到特征波长；以偏最小二乘回归（PLSR）建立各土壤养分含量的模型，通过对模型效果的评价标准（建模集绝对系数R2_c、校正均方根误差RMSEC、检验集绝对系数R2_p、预测均方根误差RMSEP和相对分析误差RPD值）来判别各单分类器算法和多分类器融合算法对土壤养分含量特征波长的提取效果。分别对四种算法、筛选其中三种算法、最优二种算法进行融合，分析融合后模型效果和特征波长个数，结果表明：将四种单分类器经投票法融合后，其模型效果大部分不如单分类器，且相对好的模型特征波长个数较多；相较于投票法多分类器融合，四种单分类器经加权投票法融合模型效果有了一定的提高，TC和TN都能够在较少的波长中获得较好的预测效果，但仅TN经融合后，模型效果优于每个单分类器；TC，TN，TP分别在取SPA+UVE+GA，SPA+UVE+GA（或SPA+GA+CC）、SPA+UVE+GA三种单分类器进行加权投票法融合后，均能获得最优模型效果，且明显优于每个单分类器，模型效果有了显著提高；各土壤养分含量经两个最优单分类器加权投票法融合后，仍能得到好于最优单分类器的建模效果，TC和TP建模效果略差于三个单分类器融合结果，TN建模效果与三个单分类器融合结果相同。因此，在筛选三种算法融合，且其中包含最优两种算法的情况下，能够以较少的特征波长个数获得明显高于单分类器的建模效果。该方法为寻找土壤养分以及其他复杂物质成分的光谱特征波段提供了新方法，也为多种算法的综合运用提供了新思路。     

基于组合支持向量机的水声目标智能识别研究

为解决水声目标小样本模式识别问题,有效地提高复杂海洋环境中的识别精度,提出了一种基于经验模式分解(EMD)、特征距离评估技术(FDET)和组合支持向量机(CSVMs)的水声目标智能识别方法。首先,将滤波、Hilbert包络解调和EMD等信号处理方法对水声目标的辐射噪声信号进行预处理,提取7个包含原始信号和预处理信号的时域和频域统计特征的特征集。然后,通过FDET从原始特征集中选择出7个敏感特征集。最后,将7个敏感特征集输入到7个支持向量机分类器中,利用遗传算法对7个分类器的结果进行合并,构成CSVMs分类器,从而实现对水声目标的智能识别。将该方法应用于舰船等水声目标的识别中,研究结果表明,该方法的识别性能优于单一SVMs分类器:同时,经过FDET得到的敏感特征集能明显地提高识别精度。     

红外热成像辅助面神经功能自动评估方法研究

面瘫是一种多发的面神经疾病,表现为患侧面神经功能失调,严重影响患者的正常生活和人际交往。面神经功能自动评估方法对于面瘫的诊治是至关重要的。面部神经功能受损导致体表温度分布发生改变,可以通过红外热成像采集患者的面部温度分布不对称特征,基于红外热成像提出一种面神经功能自动评估新方法,融合温度特异性和边缘检测自动将面部红外热像划分为左右对称的八个区域,提取面部温度不对称特征,包括温差、有效热面积比和温度分布不对称度,采用径向基神经网络作为面神经功能自动分类器。实验收录了390幅单侧患病的面瘫患者正面红外热像图,结果显示：采用径向基神经网络的红外热成像面神经功能自动分类器的平均分类准确率为94.10%,比采用K近邻分类器和支持向量机分类器分别提高了9.31%和4.87%,优于传统的House-Brackmann面神经功能评估方法,对面神经功能的分类精度完全符合临床应用标准,可以有效评估面瘫患者的面神经功能,有助于面瘫的临床诊断与治疗。     

Classifier ensembles for fMRI data analysis: an experiment

Functional magnetic resonance imaging (fMRI) is becoming a forefront brain–computer interface tool. To decipher brain patterns, fast, accurate and reliable classifier methods are needed. The support vector machine (SVM) classifier has been traditionally used. Here we argue that state-of-the-art methods from pattern recognition and machine learning, such as classifier ensembles, offer more accurate classification. This study compares 18 classification methods on a publicly available real data set due to Haxby et al. [Science 293 (2001) 2425–2430]. The data comes from a single-subject experiment, organized in 10 runs where eight classes of stimuli were presented in each run. The comparisons were carried out on voxel subsets of different sizes, selected through seven popular voxel selection methods. We found that, while SVM was robust, accurate and scalable, some classifier ensemble methods demonstrated significantly better performance. The best classifiers were found to be the random subspace ensemble of SVM classifiers, rotation forest and ensembles with random linear and random spherical oracle.     

Experimental Study and Comparison of Imbalance Ensemble Classifiers with Dynamic Selection Strategy

Imbalance ensemble classification is one of the most essential and practical strategies for improving decision performance in data analysis. There is a growing body of literature about ensemble techniques for imbalance learning in recent years, the various extensions of imbalanced classification methods were established from different points of view. The present study is initiated in an attempt to review the state-of-the-art ensemble classification algorithms for dealing with imbalanced datasets, offering a comprehensive analysis for incorporating the dynamic selection of base classifiers in classification. By conducting 14 existing ensemble algorithms incorporating a dynamic selection on 56 datasets, the experimental results reveal that the classical algorithm with a dynamic selection strategy deliver a practical way to improve the classification performance for both a binary class and multi-class imbalanced datasets. In addition, by combining patch learning with a dynamic selection ensemble classification, a patch-ensemble classification method is designed, which utilizes the misclassified samples to train patch classifiers for increasing the diversity of base classifiers. The experiments’ results indicate that the designed method has a certain potential for the performance of multi-class imbalanced classification.     

近红外光谱和多分类器融合的葡萄酒品种判别研究

将多种单分类器模型融合,并用融合后的模型对不同品种干红葡萄酒进行判别分析。用BRUKER MPA傅里叶变换型近红外光谱仪采集170个干红葡萄酒样品的近红外透射光谱,选取PLS-DA,SVM,Fisher和AdaBoost作为单分类器建模方法,分别建立葡萄酒品种判别模型,通过差异性度量值对单分类器进行筛选,得到差异性较大的四个单分类器作为基分类器,其中基分类器对测试集葡萄酒品种判别准确率最高为88.24%,最低为81.18%。然后通过加权投票机制对基分类器进行融合,融合后的模型对测试集葡萄酒品种判别准确率提高至92.94%,误判样品个数由单分类器最少的9个降为6个。实验结果表明多分类器融合所建立的模型优于传统近红外光谱定性分析一般采用单分类器模型结果,提高了葡萄酒品种判别的准确性,采用基于近红外光谱的多分类融合方法对葡萄酒种类判定具有可行性。     

Nonparametric subspace analysis fused to 2DPCA for face recognition

Two-dimensional principal component analysis (2DPCA) is one of the representative techniques for image representation and recognition. However, keen storage requirements and computational complexity consist in 2DPCA. Meanwhile, the performance of 2DPCA is delicate in illumination variations. Nonparametric subspace analysis (NSA) is a subspace learning method that can reduce dimensionality and identify local information for discrimination, so that it can make 2DPCA perform well in illumination. Motivated by above facts, 2DPCA fused with NSA is implemented for face recognition, which can reduce dimensions of the 2DPCA feature vectors and enhance the contribution of principal components to face recognition. Experiments carried out on ORL, Yale B, and FERET facial databases show that valid recognition rates can be achieved by the proposed method compared to 2DPCA, 2DPCA plus PCA, LDA methods and demonstrate promising abilities against illumination variations.     

不均匀光照的路面裂缝检测和分类新方法

针对受光照不均影响的路面裂缝图像,提出一种基于Sobel算子和最大熵法的图像分割算法,并采用长线段与原图进行与操作和判断黑色像素所占比例的方法去除图像孤立噪声点.根据不同类型裂缝的几何形态,提取投影向量、分布密度和空洞数等特征值作为路面裂缝分类的依据,设计径向基函数神经网络的分类器实现对裂缝的准确分类.实验结果表明,较传统全局阈值算法,本文算法对光照不均图像的处理不仅能很好的提取裂缝边缘,且具有很强的抗噪能力,对路面裂缝的分类准确率高.     

基于高维拉曼光谱数据的变压器油纸绝缘老化评估方法研究

采用激光拉曼光谱技术对变压器油纸绝缘老化状态检测是一种有效的方法。随着样本量的扩充,亟待处理的数据集维度逐渐增大,研究适用于高维拉曼光谱数据的变压器油纸绝缘老化评估方法具有重要的意义。设计与现场变压器内部绝缘结构相似的油纸绝缘环境,进行加速热老化实验并定期采样,获取到10类老化程度依次递增的油样本,采用激光拉曼光谱技术对样本进行检测。选用复合稀疏导数建模法对样本原始拉曼光谱数据预处理,可以一步完成去噪与基线校正;引入差异特征选取方法筛选不同老化程度下光谱中变化显著的特征,计算同一拉曼频移下不同老化程度的特征点数据集方差,选择差异较大的数据序列所对应的拉曼特征变量,设定方差阈值为0.5进行特征选择,每个样本都从1 023个光谱特征点抽取出304个特征点进行后续分析;针对变压器油纸绝缘老化拉曼光谱高维样本数据集,引入多种不同类型的算法对其处理。分别运用K-means聚类算法、Fisher算法与随机森林算法对获取到的样本预处理后的数据建立模型,引入评估准确度、提升度以及Kappa系数对各算法建立的模型判别效果进行评估。结果表明：有监督学习的Fisher算法与随机森林算法效果较好,相对于无监督学习的K-means聚类算法,模型判别能力分别提升了1.166 6和1.95,论证了有监督学习模型在变压器油纸绝缘老化的评估中具有判别优势;从模型判别准确度和Kappa系数来看,强分类器随机森林算法建立的判别模型均高于Fisher判别模型,其准确度提升了10%,且Kappa系数上升了0.111 5,论证了随机森林算法作为由多个单一分类器组成的强分类器,相对单一分类器来说,在变压器油纸绝缘老化的评估中模型的泛化能力较好,且模型较为稳定可靠。通过对三种不同类型的算法对比,确定了在变压器油纸绝缘老化评估中,有监督学习强分类器随机森林算法的判别优势,为变压器油纸绝缘老化的有效评估打下了基础。     

低照度条件下超分辨率人脸图像采集系统设计与实现

在计算机技术高速发展的时代,多平台计算机视觉库随之产生。OpenCV作为一种开源代码的计算机视觉库,以可兼容多平台、接口广泛的特点被广泛运用各个领域。在低照度条件下,会出现光照环境差异过大或光线不足等情况,导致传统图像采集系统不能采集高质量的人脸图像,局限性较差。提出基于OpenCV在C 环境配置下运用三维人脸识别技术算法,设计一套低照度条件下超分辨率人脸图像采集系统。实验证明,该设计方案具有实时(对焦速度快)、快速(单张采集0.05秒)、准确(面部识别率99.3%)等特点,能够充分满足低照度条件下超分辨率人脸图像采集的需求。     

Link Prediction in Complex Networks Using Recursive Feature Elimination and Stacking Ensemble Learning

Link prediction is an important task in the field of network analysis and modeling, and predicts missing links in current networks and new links in future networks. In order to improve the performance of link prediction, we integrate global, local, and quasi-local topological information of networks. Here, a novel stacking ensemble framework is proposed for link prediction in this paper. Our approach employs random forest-based recursive feature elimination to select relevant structural features associated with networks and constructs a two-level stacking ensemble model involving various machine learning methods for link prediction. The lower level is composed of three base classifiers, i.e., logistic regression, gradient boosting decision tree, and XGBoost, and their outputs are then integrated with an XGBoost model in the upper level. Extensive experiments were conducted on six networks. Comparison results show that the proposed method can obtain better prediction results and applicability robustness.     

融合Retinex和离散小波奇异值分解的远距离目标图像清晰化

针对远距离成像系统获取的低照度降质图像增强问题,提出了一种融合Retinex和离散小波奇异值分解的图像清晰化算法。该方法首先利用自适应全尺度Retinex（adaptive full-scale retinex, AFSR）“粗”提取照度分量和反射分量,然后通过离散小波变换将所提取的图像反射分量分解为4个频率子带并估计出低频子带图像的奇异值矩阵,最后应用逆小波变换“精”重建图像。实验结果表明:所提方法处理后的低照度降质图像视觉增强效果较好,在图像对比度、信息熵、平均梯度和边缘密度等客观评价指标方面优于其他经典算法。     

Combined multi-kernel support vector machine and wavelet analysis for hyperspectral remote sensing image classification

Many remote sensing image classifiers are limited in their ability to combine spectral features with spatial features.Multi-kernel classifiers,however,are capable of integrating spectral features with spatial or structural features using multiple kernels and summing them for final outputs.Using a support vector machine(SVM) as classifier,different multi-kernel classifiers are constructed and tested using 64-band Operational Modular Imaging Spectrometer II hyperspectral image of Changping Area,Beijing City.Results show that by integrating spectral and wavelet texture information,multi-kernel SVM classifiers can obtain more accurate classification results than sole-kernel SVM classifiers and cross-information SVM kernel classifiers.Moreover,when the multi-kernel SVM classifier is used,the combination of the first four principal components from principal component analysis and wavelet texture provides the highest accuracy(97.06%).Multi-kernel SVM is therefore an effective approach to improve the accuracy of hyperspectral image classification and to expand possibilities for remote sensing image interpretation and application.     

Unified Generative Adversarial Networks for Multidomain Fingerprint Presentation Attack Detection

With the rapid growth of fingerprint-based biometric systems, it is essential to ensure the security and reliability of the deployed algorithms. Indeed, the security vulnerability of these systems has been widely recognized. Thus, it is critical to enhance the generalization ability of fingerprint presentation attack detection (PAD) cross-sensor and cross-material settings. In this work, we propose a novel solution for addressing the case of a single source domain (sensor) with large labeled real/fake fingerprint images and multiple target domains (sensors) with only few real images obtained from different sensors. Our aim is to build a model that leverages the limited sample issues in all target domains by transferring knowledge from the source domain. To this end, we train a unified generative adversarial network (UGAN) for multidomain conversion to learn several mappings between all domains. This allows us to generate additional synthetic images for the target domains from the source domain to reduce the distribution shift between fingerprint representations. Then, we train a scale compound network (EfficientNetV2) coupled with multiple head classifiers (one classifier for each domain) using the source domain and the translated images. The outputs of these classifiers are then aggregated using an additional fusion layer with learnable weights. In the experiments, we validate the proposed methodology on the public LivDet2015 dataset. The experimental results show that the proposed method improves the average classification accuracy over twelve classification scenarios from 67.80 to 80.44% after adaptation.     

一种全局参数估计的水下主动偏振去雾算法

为改善水下主动光照明条件下后向散射光对成像的影响,通过分析水下主动偏振成像模型,提出一种基于拟合函数的全局参数估计的水下主动偏振去雾算法。该算法结合图像增强作为图像预处理,再设定二元多项式函数,利用最小二乘法进行后向散射光偏振度变量的拟合求解,得到对比度更高、信息更丰富的水下复原图像。实验结果表明:该算法可有效改善主动光照明条件下的水下图像质量,提高图像对比度,复原被淹没的细节信息,复原图像的图像增强测量值较以往算法相比提升70%,且能适用于不同浓度介质的情况。