Fuzzy Representation of Principal’s Preferences in Inspire Negotiation Support System

We consider the negotiation problem, in which an agent negotiates on behalf of a principal. Our considerations are focused on the Inspire negotiation support system in which the principal’s preferences are visualised by circles. In this way, the principal describes the importance of each negotiation issue and the relative utility of each considered option. The paper proposes how this preference information may be implemented by the agent for determining a scoring function used to support decisions throughout the negotiation process. The starting point of our considerations is a discussion regarding the visualisation of the principal’s preferences. We assume here that the importance of each issue and the utility of each option increases with the size of the circle representing them. The imprecise meaning of the notion of “circle size” implies that in a considered case, the utility of an option should be evaluated by a fuzzy number. The proposed utility fuzzification is justified by a simple analysis of results obtained from the empirical prenegotiation experiment. A novel method is proposed to determine trapezoidal fuzzy numbers, which evaluates an option’s utility using a series of answers given by the participants of the experiment. The utilities obtained this way are applied to determine the fuzzy scoring function for an agent. By determining such a common generalised fuzzy scoring system, our approach helps agents handle the differences in human cognitive processes associated with understanding the principal’s preferences. This work is the first approach to fuzzification of the preferences in the Inspire negotiation support system.     

Two-Party Privacy-Preserving Set Intersection with FHE

A two-party private set intersection allows two parties, the client and the server, to compute an intersection over their private sets, without revealing any information beyond the intersecting elements. We present a novel private set intersection protocol based on Shuhong Gao’s fully homomorphic encryption scheme and prove the security of the protocol in the semi-honest model. We also present a variant of the protocol which is a completely novel construction for computing the intersection based on Bloom filter and fully homomorphic encryption, and the protocol’s complexity is independent of the set size of the client. The security of the protocols relies on the learning with errors and ring learning with error problems. Furthermore, in the cloud with malicious adversaries, the computation of the private set intersection can be outsourced to the cloud service provider without revealing any private information.     

A Novel Conflict Management Method Based on Uncertainty of Evidence and Reinforcement Learning for Multi-Sensor Information Fusion

Dempster–Shafer theory (DST), which is widely used in information fusion, can process uncertain information without prior information; however, when the evidence to combine is highly conflicting, it may lead to counter-intuitive results. Moreover, the existing methods are not strong enough to process real-time and online conflicting evidence. In order to solve the above problems, a novel information fusion method is proposed in this paper. The proposed method combines the uncertainty of evidence and reinforcement learning (RL). Specifically, we consider two uncertainty degrees: the uncertainty of the original basic probability assignment (BPA) and the uncertainty of its negation. Then, Deng entropy is used to measure the uncertainty of BPAs. Two uncertainty degrees are considered as the condition of measuring information quality. Then, the adaptive conflict processing is performed by RL and the combination two uncertainty degrees. The next step is to compute Dempster’s combination rule (DCR) to achieve multi-sensor information fusion. Finally, a decision scheme based on correlation coefficient is used to make the decision. The proposed method not only realizes adaptive conflict evidence management, but also improves the accuracy of multi-sensor information fusion and reduces information loss. Numerical examples verify the effectiveness of the proposed method.     

A Machine Learning Approach for Gearbox System Fault Diagnosis

This study proposes a fully automated gearbox fault diagnosis approach that does not require knowledge about the specific gearbox construction and its load. The proposed approach is based on evaluating an adaptive filter’s prediction error. The obtained prediction error’s standard deviation is further processed with a support-vector machine to classify the gearbox’s condition. The proposed method was cross-validated on a public dataset, segmented into 1760 test samples, against two other reference methods. The accuracy achieved by the proposed method was better than the accuracies of the reference methods. The accuracy of the proposed method was on average 9% higher compared to both reference methods for different support vector settings.     

Hierarchical Classification of Event-Related Potentials for the Recognition of Gender Differences in the Attention Task

Research on the functioning of human cognition has been a crucial problem studied for years. Electroencephalography (EEG) classification methods may serve as a precious tool for understanding the temporal dynamics of human brain activity, and the purpose of such an approach is to increase the statistical power of the differences between conditions that are too weak to be detected using standard EEG methods. Following that line of research, in this paper, we focus on recognizing gender differences in the functioning of the human brain in the attention task. For that purpose, we gathered, analyzed, and finally classified event-related potentials (ERPs). We propose a hierarchical approach, in which the electrophysiological signal preprocessing is combined with the classification method, enriched with a segmentation step, which creates a full line of electrophysiological signal classification during an attention task. This approach allowed us to detect differences between men and women in the P3 waveform, an ERP component related to attention, which were not observed using standard ERP analysis. The results provide evidence for the high effectiveness of the proposed method, which outperformed a traditional statistical analysis approach. This is a step towards understanding neuronal differences between men’s and women’s brains during cognition, aiming to reduce the misdiagnosis and adverse side effects in underrepresented women groups in health and biomedical research.     

基于可见光谱和卷积神经网络的贝类识别方法

目前卷积神经网络（CNN）在物体种类识别方面取得突破性进展。贝类作为农业经济的重要组成部分,种类繁多,特点复杂,大多贝类存在着相似度高,各类样本分布不均衡情况,以致CNN对贝类分类的准确率偏低。针对这一情况,提出了基于可见光谱和CNN的贝类识别方法,旨在提取更有效的贝类特征,从而提高贝类分类的准确率。首先,提出了一种包含输出熵度量和正交性度量的滤波器信息度量与特征选择方法,重新初始化修剪掉的滤波器并使其正交,捕获网络激活空间中的不同方向,使神经网络模型学习到更多有用的贝类特征信息,提升模型分类准确率;其次,提出了一种包含正则化项和焦点损失项的贝类分类目标函数,通过控制各类别样本对总损失的共享权重,来减少易分类样本的权重,以使模型注意力向预测不准的样本倾斜,均衡样本分布和样本分类难度,进一步提高贝类分类的准确率。贝类图像数据集由74类贝类组成,共11 803张图像。获取原始数据集后,对数据集图像进行水平翻转、垂直翻转、随机旋转、在[0, 30°]范围内旋转、在[0, 20%]范围内缩放和移动等数据增强操作,将图像数量从11 803张增加到119 964张。整个图像数据集按8∶1∶1的比例随机分为训练集95 947张图片、验证集11 996张图片和测试集12 021张图片。在建立贝类图像数据集的基础上进行了实验验证,达到了93.38%的分类准确率,将基准网络（Resnest）的准确率提高了1.18%,相较网络SN_Net和MutualNet,准确率分别提升了4.34%和0.85% ,并且训练时长为22 320 s,将基准网络（Resnest）的训练时长缩短了960 s,训练时长分别比SN_Net和MutualNet短3 180和2 460 s。实验结果证明了该方法的有效性。     

Analysis of Electromagnetic Information Leakage Based on Cryptographic Integrated Circuits

Cryptographic algorithm is the most commonly used method of information security protection for many devices. The secret key of cryptographic algorithm is usually stored in these devices’ registers. In this paper, we propose an electromagnetic information leakage model to investigate the relationship between the electromagnetic leakage signal and the secret key. The registers are considered as electric dipole models to illustrate the source of the electromagnetic leakage. The equivalent circuit of the magnetic field probe is developed to bridge the output voltage and the electromagnetic leakage signal. Combining them, the electromagnetic information leakage model’s function relationship can be established. Besides, an electromagnetic leakage model based on multiple linear regression is proposed to recover the secret key and the model’s effectiveness is evaluated by guess entropy. Near field tests are conducted in an unshielded ordinary indoor environment to investigate the electromagnetic side-channel information leakage. The experiment result shows the correctness of the proposed electromagnetic leakage model and it can be used to recover the secret key of the cryptographic algorithm.     

A Multi-Objective Multi-Label Feature Selection Algorithm Based on Shapley Value

Multi-label learning is dedicated to learning functions so that each sample is labeled with a true label set. With the increase of data knowledge, the feature dimensionality is increasing. However, high-dimensional information may contain noisy data, making the process of multi-label learning difficult. Feature selection is a technical approach that can effectively reduce the data dimension. In the study of feature selection, the multi-objective optimization algorithm has shown an excellent global optimization performance. The Pareto relationship can handle contradictory objectives in the multi-objective problem well. Therefore, a Shapley value-fused feature selection algorithm for multi-label learning (SHAPFS-ML) is proposed. The method takes multi-label criteria as the optimization objectives and the proposed crossover and mutation operators based on Shapley value are conducive to identifying relevant, redundant and irrelevant features. The comparison of experimental results on real-world datasets reveals that SHAPFS-ML is an effective feature selection method for multi-label classification, which can reduce the classification algorithm’s computational complexity and improve the classification accuracy.     

Speech emotion recognition based on statistical pitch model

A modified Parzen-window method, which keep high resolution in low frequencies and keep smoothness in high frequencies, is proposed to obtain statistical model. Then, a gender classification method utilizing the statistical model is proposed, which have a 98% accuracy of gender classification while long sentence is dealt with. By separation the male voice and female voice, the mean and standard deviation of speech training samples with different emotion are used to create the corresponding emotion models. Then the Bhattacharyya distance between the test sample and statistical models of pitch, are utilized for emotion recognition in speech. The normalization of pitch for the male voice and female voice are also considered, in order to illustrate them into a uniform space. Finally, the speech emotion recognition experiment based on K Nearest Neighbor shows that, the correct rate of 81% is achieved, where it is only 73.85% if the traditional parameters are utilized.     

Robust pattern decoding in shape-coded structured light

Decoding is a challenging and complex problem in a coded structured light system. In this paper, a robust pattern decoding method is proposed for the shape-coded structured light in which the pattern is designed as grid shape with embedded geometrical shapes. In our decoding method, advancements are made at three steps. First, a multi-template feature detection algorithm is introduced to detect the feature point which is the intersection of each two orthogonal grid-lines. Second, pattern element identification is modelled as a supervised classification problem and the deep neural network technique is applied for the accurate classification of pattern elements. Before that, a training dataset is established, which contains a mass of pattern elements with various blurring and distortions. Third, an error correction mechanism based on epipolar constraint, coplanarity constraint and topological constraint is presented to reduce the false matches. In the experiments, several complex objects including human hand are chosen to test the accuracy and robustness of the proposed method. The experimental results show that our decoding method not only has high decoding accuracy, but also owns strong robustness to surface color and complex textures.     

Development of a Neurodegenerative Disease Gait Classification Algorithm Using Multiscale Sample Entropy and Machine Learning Classifiers

The prevalence of neurodegenerative diseases (NDD) has grown rapidly in recent years and NDD screening receives much attention. NDD could cause gait abnormalities so that to screen NDD using gait signal is feasible. The research aim of this study is to develop an NDD classification algorithm via gait force (GF) using multiscale sample entropy (MSE) and machine learning models. The Physionet NDD gait database is utilized to validate the proposed algorithm. In the preprocessing stage of the proposed algorithm, new signals were generated by taking one and two times of differential on GF and are divided into various time windows (10/20/30/60-sec). In feature extraction, the GF signal is used to calculate statistical and MSE values. Owing to the imbalanced nature of the Physionet NDD gait database, the synthetic minority oversampling technique (SMOTE) was used to rebalance data of each class. Support vector machine (SVM) and k-nearest neighbors (KNN) were used as the classifiers. The best classification accuracies for the healthy controls (HC) vs. Parkinson’s disease (PD), HC vs. Huntington’s disease (HD), HC vs. amyotrophic lateral sclerosis (ALS), PD vs. HD, PD vs. ALS, HD vs. ALS, HC vs. PD vs. HD vs. ALS, were 99.90%, 99.80%, 100%, 99.75%, 99.90%, 99.55%, and 99.68% under 10-sec time window with KNN. This study successfully developed an NDD gait classification based on MSE and machine learning classifiers.     

近景大视场角分区域标定方法研究

针对飞行试验测量视场大相机标定精度低的问题,提出一种高精度CCD相机分区域标定方法。该方法首先通过将标靶均匀布置在摄像机视场内,使得标靶尽可能均匀错落地充满整个视场范围,再结合人眼判读的方式求解靶标的像面位置,最终与全站仪三维坐标形成精确的空间标定点集。接着,将像平面按横向方向等间距分割成Ｎ个区域,并结合后方交会的方法分别对每个子区域进行相机参数的计算。实验结果表明：经过分区域标定,相机采集点的总误差比单区域标定法降低了4％(Ｎ＝3)。算法可实现指定区域的相机参数计算,基本满足中高等精度的工业测量要求。所本文研究可应用于位置相对固定不变的工业视觉测量,特别是大工件测量领域。     

Real Sample Consistency Regularization for GANs

Mode collapse has always been a fundamental problem in generative adversarial networks. The recently proposed Zero Gradient Penalty (0GP) regularization can alleviate the mode collapse, but it will exacerbate a discriminator’s misjudgment problem, that is the discriminator judges that some generated samples are more real than real samples. In actual training, the discriminator will direct the generated samples to point to samples with higher discriminator outputs. The serious misjudgment problem of the discriminator will cause the generator to generate unnatural images and reduce the quality of the generation. This paper proposes Real Sample Consistency (RSC) regularization. In the training process, we randomly divided the samples into two parts and minimized the loss of the discriminator’s outputs corresponding to these two parts, forcing the discriminator to output the same value for all real samples. We analyzed the effectiveness of our method. The experimental results showed that our method can alleviate the discriminator’s misjudgment and perform better with a more stable training process than 0GP regularization. Our real sample consistency regularization improved the FID score for the conditional generation of Fake-As-Real GAN (FARGAN) from 14.28 to 9.8 on CIFAR-10. Our RSC regularization improved the FID score from 23.42 to 17.14 on CIFAR-100 and from 53.79 to 46.92 on ImageNet2012. Our RSC regularization improved the average distance between the generated and real samples from 0.028 to 0.025 on synthetic data. The loss of the generator and discriminator in standard GAN with our regularization was close to the theoretical loss and kept stable during the training process.     

基于克隆选择支持向量机高光谱遥感影像分类技术

作为支持向量机(support vector machine, SVM)高光谱影像分类的一个重要环节,参数设置的效率和精度直接影响到SVM模型训练效率和最终分类精度。本文首先建立一个SVM高光谱影像分类器,提出了利用免疫克隆选择算法优化的交叉验证进行核函数参数和惩罚因子C的优化选择的方法,得到了一种基于克隆选择优化的支持向量机(clonal selection SVM, CSSVM)高光谱影像分类器。然后将CSSVM与传统的基于网格搜索交叉验证的支持向量机（gird search SVM, GSSVM）分类器进行了对比评价,评价指标包括模型训练时间和分类精度等。最后基于AVIRIS高光谱遥感影像进行了两算法分类对比试验,结果表明：提出的CSSVM测试样本总分类精度超过85.1%和Kappa系数超过0.821 3,影像总分类精度超过81.58%和Kappa系数超过0.772 8,CSSVM与GSSVM的分类精度差别在0.08%以内,Kappa系数差别在0.001以内;CSSVM的模型训练时间是GSSVM的1/6至1/10,得到显著缩短;CSSVM方法在保持传统GSSVM优良分类精度的基础上,极大提高了模型的训练效率。     

Enhancing convolutional neural network scheme for rheumatoid arthritis grading with limited clinical data

The gray-scale ultrasound(US) imaging method is usually used to assess synovitis in rheumatoid arthritis(RA) in clinical practice. This four-grade scoring system depends highly on the sonographer's experience and has relatively lower validity compared with quantitative indexes. However, the training of a qualified sonographer is expensive and timeconsuming while few studies focused on automatic RA grading methods. The purpose of this study is to propose an automatic RA grading method using deep convolutional neural networks(DCNN) to assist clinical assessment. Gray-scale ultrasound images of finger joints are taken as inputs while the output is the corresponding RA grading results. Firstly,we performed the auto-localization of synovium in the RA image and obtained a high precision in localization. In order to make up for the lack of a large annotated training dataset, we performed data augmentation to increase the number of training samples. Motivated by the approach of transfer learning, we pre-trained the GoogLeNet on ImageNet as a feature extractor and then fine-tuned it on our own dataset. The detection results showed an average precision exceeding 90%. In the experiment of grading RA severity, the four-grade classification accuracy exceeded 90% while the binary classification accuracies exceeded 95%. The results demonstrate that our proposed method achieves performances comparable to RA experts in multi-class classification. The promising results of our proposed DCNN-based RA grading method can have the ability to provide an objective and accurate reference to assist RA diagnosis and the training of sonographers.     

半监督判别分析的跨库语音情感识别

针对训练样本与测试样本来自不同语音情感数据库造成特征向量空间分布不匹配的问题,采用半监督判别分析减小二者的差异。首先寻找有标签的训练样本和来自另一个库的部分无标签训练样本之间的最优投影方向。基于一致性假设即相近的点更有可能具有相同的类别,利用p近邻图对无标签训练样本相近点之间的关系进行建模,从而获得无标签样本的分布信息。在保证无标签样本间流形结构的同时,使所有训练样本类间散度和类内散度的比值达到最大,从而得到最优的投影方向。采用两组实验进行验证,第1组用eNTERFACE库训练去测试Berlin库,识别率为51.41%,第2组用Berlin库训练测试eNTERFACE库,识别率为45.76%,相比未采用半监督判别分析的识别结果分别有了13.72%和22.81%的提高,说明该算法的有效性。通过实验前后数据的可视化分析,说明利用半监督判别分析确实减小了不同库之间特征向量空间分布的不匹配问题,从而提高跨库语音情感识别率。      

Joint UAVs’ Load Balancing and UEs’ Data Rate Fairness Optimization by Diffusion UAV Deployment Algorithm in Multi-UAV Networks

Unmanned aerial vehicles (UAVs) can be deployed as base stations (BSs) for emergency communications of user equipments (UEs) in 5G/6G networks. In multi-UAV communication networks, UAVs’ load balancing and UEs’ data rate fairness are two challenging problems and can be optimized by UAV deployment strategies. In this work, we found that these two problems are related by the same performance metric, which makes it possible to optimize the two problems simultaneously. To solve this joint optimization problem, we propose a UAV diffusion deployment algorithm based on the virtual force field method. Firstly, according to the unique performance metric, we define two new virtual forces, which are the UAV-UAV force and UE-UAV force defined by FU and FV, respectively. FV is the main contributor to load balancing and UEs’ data rate fairness, and FU contributes to fine tuning the UEs’ data rate fairness performance. Secondly, we propose a diffusion control stratedy to the update UAV-UAV force, which optimizes FV in a distributed manner. In this diffusion strategy, each UAV optimizes the local parameter by exchanging information with neighbor UAVs, which achieve global load balancing in a distributed manner. Thirdly, we adopt the successive convex optimization method to update FU, which is a non-convex problem. The resultant force of FV and FU is used to control the UAVs’ motion. Simulation results show that the proposed algorithm outperforms the baseline algorithm on UAVs’ load balancing and UEs’ data rate fairness.     

基于卷积神经网络的恒星光谱自动分类方法

恒星光谱自动分类是研究恒星光谱的基础内容,快速、准确自动识别、分类恒星光谱可提高搜寻特殊天体速度,对天文学研究有重大意义。目前我国大型巡天项目LAMOST每年发布数百万条光谱数据,对海量恒星光谱进行快速、准确自动识别与分类研究已成为天文学大数据分析与处理领域的研究热点之一。针对恒星光谱自动分类问题,提出一种基于卷积神经网络(CNN)的K和F型恒星光谱分类方法,并与支持向量机(SVM)、误差反向传播算法(BP)对比,采用交叉验证方法验证分类器性能。与传统方法相比CNN具有权值共享,减少模型学习参数;可直接对训练数据自动进行特征提取等优点。实验采用Tensorflow深度学习框架,Python3.5编程环境。K和F恒星光谱数据集采用国家天文台提供的LAMOST DR3数据。截取每条光谱波长范围为3 500~7 500 部分,对光谱均匀采样生成数据集样本,采用min-max归一化方法对数据集样本进行归一化处理。CNN结构包括：输入层,卷积层C1,池化层S1,卷积层C2,池化层S2,卷积层C3,池化层S3,全连接层,输出层。输入层为一批K和F型恒星光谱相同的3 700个波长点处流量值。C1层设有10个大小为1×3步长为1的卷积核。S1层采用最大池化方法,采样窗口大小为1×2,无重叠采样,生成10张特征图,与C1层特征图数量相同,大小为C1层特征图的二分之一。C2层设有20个大小为1×2步长为1的卷积核,输出20张特征图。S2层对C2层20张特征图下采样输出20张特征图。C3层设有30个大小为1×3步长为1的卷积核,输出30张特征图。S3层对C3层30张特征图下采样输出30张特征图。全连接层神经元个数设置为50,每个神经元都与S3层的所有神经元连接。输出层神经元个数设置为2,输出分类结果。卷积层激活函数采用ReLU函数,输出层激活函数采用softmax函数。对比算法SVM类型为C-SVC,核函数采用径向基函数,BP算法设有3个隐藏层,每个隐藏层设有20,40和20个神经元。数据集分为训练数据和测试数据,将训练数据的40%,60%,80%和100%作为5个训练集,测试数据作为测试集。分别将5个训练集放入模型中训练,共迭代8 000次,每次训练好的模型用测试集进行验证。对比实验采用100%的训练数据作为训练集,测试数据作为测试集。采用精确率、召回率、F-score、准确率四个评价指标评价模型性能,对实验结果进行详细分析。分析结果表明CNN算法可对K和F型恒星光谱快速自动分类和筛选,训练集数据量越大,模型泛化能力越强,分类准确率越高。对比实验结果表明采用CNN算法对K和F型恒星光谱自动分类较传统机器学习SVM和BP算法自动分类准确率更高。     

基于图像检索的大豆食心虫虫害高光谱检测

为减轻虫害对大豆的影响,首先使用相应的高光谱仪器进行样本采集,样本分为4类：包括带有微小虫卵的,带有幼虫的,有啃食痕迹的和完全正常的大豆各20颗;然后提出了一种基于三维图像检索(3D-R-D,3D Resnet18 DCH)的大豆食心虫的高光谱检测方法。该方法从视频检索的应用中得到启发,考虑到视频不同帧之间和高光谱不同层之间存在类比关系,使用了在大规模视频检索数据集下训练而成的分类模型,将它作为预训练3D卷积模型进行训练。和已知的文献方法相同,使用公开的光谱数据集进行正式训练和微调,从而得到能进行特征提取的3D卷积网络,用图像检索来实现间接分类,通过利用样本之间的特征距离,实现在全新类别上的分类。为能适应任务,将模型最后的分类层变成了图像检索常用的hash层,从而得到了代表特征的二进制码。该方法不但完成了对不同情况下大豆种类的检测,还解决了训练时样本不足的问题。为探寻一种好的相似度匹配损失函数,对比了多种较新的方法,最后发现使用融入柯西分布的损失函数,实验效果最佳,最终模型的分类精度达86%±1.00%,和在大豆食心虫检测上最新的小样本方法对比,3D-R-D方法提高了3.5%左右的精度,表明该方法是有效的,它也为结合高光谱检测相关研究提供了一种全新思路。     

On-Road Detection of Driver Fatigue and Drowsiness during Medium-Distance Journeys

Background: The detection of driver fatigue as a cause of sleepiness is a key technology capable of preventing fatal accidents. This research uses a fatigue-related sleepiness detection algorithm based on the analysis of the pulse rate variability generated by the heartbeat and validates the proposed method by comparing it with an objective indicator of sleepiness (PERCLOS). Methods: changes in alert conditions affect the autonomic nervous system (ANS) and therefore heart rate variability (HRV), modulated in the form of a wave and monitored to detect long-term changes in the driver’s condition using real-time control. Results: the performance of the algorithm was evaluated through an experiment carried out in a road vehicle. In this experiment, data was recorded by three participants during different driving sessions and their conditions of fatigue and sleepiness were documented on both a subjective and objective basis. The validation of the results through PERCLOS showed a 63% adherence to the experimental findings. Conclusions: the present study confirms the possibility of continuously monitoring the driver’s status through the detection of the activation/deactivation states of the ANS based on HRV. The proposed method can help prevent accidents caused by drowsiness while driving.