基于半监督动态深度融合神经网络的软测量

半监督深度神经网络建模方法已被广泛应用于软测量,但基于分层训练的网络在特征提取过程局限于挖掘每层输入的有效信息,忽略了原始输入有效信息的丢失,逐层累积,从而导致原始输入的特征表示准确率低下;另外,缺乏挖掘过程时空相关性,也会导致模型性能退化。提出一种半监督动态深度融合神经网络（semisupervised dynamics deep fusion neural network,SS-DDFNN）方法。该方法在特征提取网络的每层都重构原始输入数据并预测质量变量,通过在预训练损失中使用重构原始输入误差,减小原始输入有效信息的丢失;同时融入注意力机制和t分布随机邻域嵌入提取时空相关信息,应用提取的特征建立门控神经网络质量预测模型。实验结果显示,相较于SAE、GSTAE和SIAE模型,所提方法在脱丁烷塔案例中的预测精度分别提升了2.8%、1.1%和0.9%;在工业聚乙烯生产案例中,分别提升了2.7%、1.0%和0.7%。实验结果验证了所提方法的有效性。     

基于深度神经决策森林的体域网数据融合方法

针对体域网的多传感器数据采集过程中存在的数据冗余大、特征信息模糊问题,提出了一种基于深度神经决策森林（DNDF）的数据融合方法。首先根据目标任务的实际需求,使用卷积神经网络进行相关特征提取,再将决策树放置到全连接层之后进行精细化数据分类。通过使用DNDF方法,不仅能够有效提取多维数据的关键特征,而且能够较好地兼顾数据间的关联性。实验以AReM数据集作为实验样本,结果表明,DNDF方法相对其他传统算法具有更好的分类准确率,分类准确率达到了96.5%。     

Chinese medical question answer selection via hybrid models based on CNN and GRU

Question answer selection in the Chinese medical field is very challenging since it requires effective text representations to capture the complex semantic relationships between Chinese questions and answers. Recent approaches on deep learning, e.g., CNN and RNN, have shown their potential in improving the selection quality. However, these existing methods can only capture a part or one-side of semantic relationships while ignoring the other rich and sophisticated ones, leading to limited performance improvement. In this paper, a series of neural network models are proposed to address Chinese medical question answer selection issue. In order to model the complex relationships between questions and answers, we develop both single and hybrid models with CNN and GRU to combine the merits of different neural network architectures. This is different from existing works that can onpy capture partial relationships by utilizing a single network structure. Extensive experimental results on cMedQA dataset demonstrate that the proposed hybrid models, especially BiGRU-CNN, significantly outperform the state-of-the-art methods. The source codes of our models are available in the GitHub (https://github.com/zhangyuteng/MedicalQA-CNN-BiGRU).

     

Deep and shallow features fusion based on deep convolutional neural network for speech emotion recognition

Recent years have witnessed the great progress for speech emotion recognition using deep convolutional neural networks (DCNNs). In order to improve the performance of speech emotion recognition, a novel feature fusion method is proposed. With going deeper of the convolutional layers, the convolutional feature of traditional DCNNs gradually become more abstract, which may not be the best feature for speech emotion recognition. On the other hand, the shallow feature includes only global information without the detailed information extracted by deeper convolutional layers. According to these observations, we design a deep and shallow feature fusion convolutional network, which combines the feature from different levels of network for speech emotion recognition. The proposed network allows us to fully exploit deep and shallow feature. The popular Berlin data set is used in our experiments, the experimental results show that our proposed network can further improve speech emotion recognition rate which demonstrates the effectiveness of the proposed network.     

基于深度神经网络的特征加权融合人脸识别方法

针对目前难以提取到适合用于分类的人脸特征以及在非限条件下进行人脸识别准确率低的问题,提出了一种基于深度神经网络的特征加权融合人脸识别方法(DLWF)。首先,应用主动形状模型(ASM)提取出人脸面部的主要特征点,并根据主要特征点对人脸不同器官区域进行采样;然后,将所得采样块分别输入到对应的深度信念网络(DBN)中进行训练,获得网络最优参数;最后,利用Softmax回归求出各个区域的相似度向量,将多区域的相似度向量加权融合得到综合相似度评分进行人脸识别。经ORL和WFL人脸库上进行实验验证,DLWF算法的识别准确率分别达到97%和88.76%,与传统算法主成分分析(PCA)、支持向量机(SVM)、DBN及FIP+线性判别式分析(LDA)相比,无论是限制条件还是非限制条件下,识别率均有提高。实验结果表明,该算法具有高效的人脸识别能力。     

Multi-focus image fusion with a deep convolutional neural network

As is well known, activity level measurement and fusion rule are two crucial factors in image fusion. For most existing fusion methods, either in spatial domain or in a transform domain like wavelet, the activity level measurement is essentially implemented by designing local filters to extract high-frequency details, and the calculated clarity information of different source images are then compared using some elaborately designed rules to obtain a clarity/focus map. Consequently, the focus map contains the integrated clarity information, which is of great significance to various image fusion issues, such as multi-focus image fusion, multi-modal image fusion, etc. However, in order to achieve a satisfactory fusion performance, these two tasks are usually difficult to finish. In this study, we address this problem with a deep learning approach, aiming to learn a direct mapping between source images and focus map. To this end, a deep convolutional neural network (CNN) trained by high-quality image patches and their blurred versions is adopted to encode the mapping. The main novelty of this idea is that the activity level measurement and fusion rule can be jointly generated through learning a CNN model, which overcomes the difficulty faced by the existing fusion methods. Based on the above idea, a new multi-focus image fusion method is primarily proposed in this paper. Experimental results demonstrate that the proposed method can obtain state-of-the-art fusion performance in terms of both visual quality and objective assessment. The computational speed of the proposed method using parallel computing is fast enough for practical usage. The potential of the learned CNN model for some other-type image fusion issues is also briefly exhibited in the experiments.     

Road segmentation with image-LiDAR data fusion in deep neural network

Multimedia Tools and Applications - Robust road segmentation is a key challenge in self-driving research. Though many image based methods have been studied and high performances in dataset...     

An intelligent fault diagnosis method for rotating machinery based on data fusion and deep residual neural network

Applied Intelligence - Rotating machinery is a very important mechanical device widely used in critical industrial applications. Efficient fault detection and diagnosis are key challenges in the...     

A deep neural network based classifier for brain tumor diagnosis

Classification process plays a key role in diagnosing brain tumors. Earlier research works are intended for identifying brain tumors using different classification techniques. However, the False Alarm Rates (FARs) of existing classification techniques are high. To improve the early-stage brain tumor diagnosis via classification the Weighted Correlation Feature Selection Based Iterative Bayesian Multivariate Deep Neural Learning (WCFS-IBMDNL) technique is proposed in this work. The WCFS-IBMDNL algorithm considers medical dataset for classifying the brain tumor diagnosis at an early stage. At first, the WCFS-IBMDNL technique performs Weighted Correlation-Based Feature Selection (WC-FS) by selecting subsets of medical features that are relevant for classification of brain tumors. After completing the feature selection process, the WCFS-IBMDNL technique uses Iterative Bayesian Multivariate Deep Neural Network (IBMDNN) classifier for reducing the misclassification error rate of brain tumor identification. The WCFS-IBMDNL technique was evaluated in JAVA language using Disease Diagnosis Rate (DDR), Disease Diagnosis Time (DDT), and FAR parameter through the epileptic seizure recognition dataset.     

基于神经网络的多元特征融合身份识别系统

从信息融合的角度出发,利用神经网络的方法将语音信号、人脸图像等多元特征数据信息有机地结合起来,设计并实现了一种基于神经网络的智能融合身份识别系统。系统利用改进的线性预测和轮廓检测等方法,求出3类特征参数,并将它们进行关联,最后利用RBF人工神经网络进行融合识别。实验表明,与传统的单一特征识别系统相比,该方法具有更好的识别效果。     

基于深度卷积神经网络的行人检测

行人检测一直是目标检测研究与应用中的热点。目前行人检测主要通过设计有效的特征提取方法建立对行人特征的描述,然后利用分类器实现二分类。卷积神经网络作为深度学习的重要组成,在图像、语音等领域得到了成功应用。针对人工设计的特征提取方法难以有效表达复杂环境下行人特征的问题,提出采用多层网络构建深度卷积神经网络实现对行人检测的方法。系统分析了卷积神经网络层数、卷积核大小、特征维数等对识别效果的影响,优化了网络参数。实验结果表明该方法对于行人检测具有很高的识别率,优于传统方法。     

基于深度神经网络的蒙古语声学模型建模研究

针对高斯混合模型在蒙古语语音识别声学建模中不能充分描述蒙古语声学特征之间相关性和独立性假设的问题,开展了使用深度神经网络模型进行蒙古语声学模型建模的研究。以深度神经网络为基础,将分类与语音特征内在结构的学习紧密结合进行蒙古语声学特征的提取,构建了DNN-HMM蒙古语声学模型,结合无监督预训练与监督训练调优过程设计了训练算法,在DNN-HMM蒙古语声学模型训练中加入dropout技术避免过拟合现象。最后,在小规模语料库和Kaldi实验平台下,对GMM-HMM和DNN-HMM蒙古语声学模型进行了对比实验。实验结果表明,DNN-HMM蒙古语声学模型的词识别错误率降低了7.5%,句识别错误率降低了13.63%;同时,训练时加入dropout技术可以有效避免DNN-HMM蒙古语声学模型的过拟合现象。     

基于卷积神经网络的显著性检测

显著性检测在图像处理领域应用广泛,当前显著性检测主要有自底而上与自顶而下及一些相关或改进算法,它们各有优势和缺陷。提出了一种基于卷积神经网络的显著性检测算法,利用卷积神经网络在图像处理方面强大的功能提取图像特征,进行特征融合,最后得到显著性图,用于显著性检测。将本文方法与传统的显著性检测方法进行对比,发现本文方法效果显著。     

深度卷积神经网络的显著性检测

目的 显著性检测问题是近年来的研究热点之一,针对许多传统方法都存在着特征学习不足和鲁棒检测效果不好等问题,提出一种新的基于深度卷积神经网络的显著性检测模型.方法 首先,利用超像素的方法聚类相似特征的像素点,仿人脑视皮层细胞提取目标边缘,得到区域和边缘特征.然后,通过深度卷积神经网络学习图像的区域与边缘特征,获取相应的目标检测显著度置信图.最后,将深度卷积神经网络输出的置信度融入到条件随机场,求取能量最小化,实现显著性与非显著性判别,完成显著性检测任务.结果 在两个常用的视觉检测数据库上进行实验,本文算法的检测精度与当前最好的方法相比,在MSAR数据库上检测精度相对提升大约1.5%,在Berkeley数据库上提升效果更加明显,达到了5%.此外,无论是自然场景还是人工建筑场景、大目标与小目标,检测的效果都是最好的.结论 本文融合多特征的深度学习方法与单一浅层人工特征的方法相比更有优势,它避免了手工标定特征所带来的不确定性,具有更好的鲁棒性与普适性,从主观视觉愉悦度和客观检测准确度两方面说明了算法的有效性.     

基于深度卷积神经网络的图像检索算法研究

为解决卷积神经网络在提取图像特征时所造成的特征信息损失,提高图像检索的准确率,提出了一种基于改进卷积神经网络LeNet-L的图像检索算法。首先,改进LeNet-5卷积神经网络结构,增加网络结构深度。然后,对深度卷积神经网络模型LeNet-L进行预训练,得到训练好的网络模型,进而提取出图像高层语义特征。最后,通过距离函数比较待检图像与图像库的相似度,得出相似图像。在Corel数据集上,与原模型以及传统的SVM主动学习图像检索方法相比,该图像检索方法有较高的准确性。经实验结果表明,改进后的卷积神经网络具有更好的检索效果。     

基于Transformer和双重注意力融合的分层交互答案选择模型

答案选择是问答系统中的关键组成部分,提升其准确性是问答系统研究的重要内容之一。近年来深度学习技术广泛应用于答案选择,获得了良好效果,但仍旧有一定的局限性。其中模型对问题语义信息利用度低、缺乏局部语义重视、句子间交互感知能力差等问题尤为突出。针对上述问题提出了一种基于Transformer和双重注意力融合的答案选择模型NHITAS（new hierarchical interactive Transformer for answer selection）。首先,在信息预处理部分提取问题类别和关键词对答案进行初步筛选,并引入外部知识作为句子额外语义补充;其次,采用分层交互的Transformer对句子进行语义上下文建模,提出了UP-Transformer（untied position-Transformer）和DA-Transformer（decay self-attention-Transformer）两种结构学习语义特征;最后,通过双重注意力融合过滤句子噪声,增强问题和答案之间的语义交互。在WikiQA和TrecQA数据集上对NHITAS的性能进行测试,实验结果表明所提模型对比其他模型,能够有效提升答案选择任务的效果。     

改进的弹性网模型在深度神经网络中的应用

由于具有较高的模型复杂度，深层神经网络容易产生过拟合问题，为了减少该问题对网络性能的不利影响，提出一种基于改进的弹性网模型的深度学习优化方法。首先，考虑到变量之间的相关性，对弹性网模型中的L1范数的不同变量进行自适应加权，从而得到L2范数与自适应加权的L1范数的线性组合。其次，将改进的弹性网络模型与深度学习的优化模型相结合，给出在这种新正则项约束下求解神经网络参数的过程。然后，推导出改进的弹性网模型在神经网络优化中具有群组选择能力和Oracle性质，进而从理论上保证该模型是一种更加鲁棒的正则化方法。最后，在多个回归问题和分类问题的实验中，相对于L1、L2和弹性网正则项，该方法的回归测试误差可分别平均降低87.09、88.54和47.02，分类测试准确度可分别平均提高3.98、2.92和3.58个百分点。由此，在理论和实验两方面验证了改进的弹性网模型可以有效地增强深层神经网络的泛化能力，提升优化算法的性能，解决深度学习的过拟合问题。     

卷积神经网络的多字体汉字识别

目的 多字体的汉字识别在中文自动处理及智能输入等方面具有广阔的应用前景,是模式识别领域的一个重要课题。近年来,随着深度学习新技术的出现,基于深度卷积神经网络的汉字识别在方法和性能上得到了突破性的进展。然而现有方法存在样本需求量大、训练时间长、调参难度大等问题,针对大类别的汉字识别很难达到最佳效果。方法 针对无遮挡的印刷及手写体汉字图像,提出了一种端对端的深度卷积神经网络模型。不考虑附加层,该网络主要由3个卷积层、2个池化层、1个全连接层和一个Softmax回归层组成。为解决样本量不足的问题,提出了综合运用波纹扭曲、平移、旋转、缩放的数据扩增方法。为了解决深度神经网络参数调整难度大、训练时间长的问题,提出了对样本进行批标准化以及采用多种优化方法相结合精调网络等策略。结果 实验采用该深度模型对国标一级3 755类汉字进行识别,最终识别准确率达到98.336%。同时通过多组对比实验,验证了所提出的各种方法对改善模型最终效果的贡献。其中使用数据扩增、使用混合优化方法和使用批标准化后模型对测试样本的识别率分别提高了8.0%、0.3%和1.4%。结论 与其他文献中利用手工提取特征结合卷积神经网络的方法相比,减少了人工提取特征的工作量;与经典卷积神经网络相比,该网络特征提取能力更强,识别率更高,训练时间更短。     

基于粗神经网络的仿人智能机器人的语音融合算法研究

将粗集合理论与神经网络相结合，提出一种基于粗神经网络的新的信息融合方法，用于仿人智能机器人的语音融合。该方法不仅可以接受定量输入，而且可以接受定性输入，即输入是一个范围，或在观测时间内输入是变化的。由于粗神经网络的误差传递函数不可微，所以采用遗传算法来训练粗神经网络。仿真实验结果表明，基于粗神经网络的信息融合方法有效地提高了语音的识别率。