基于深度迁移学习的微型细粒度图像分类

现有的细粒度分类模型不仅利用图像的类别标签,还使用大量人工标注的额外信息。为解决该问题,本文提出一种深度迁移学习模型,将大规模有标签细粒度数据集上学习到的图像特征有效地迁移至微型细粒度数据集中。首先,通过衔接域定量计算域间任务的关联度。然后,根据关联度选择适合目标域的迁移特征。最后,使用细粒度数据集视图类标签进行辅助学习,通过联合学习所有属性来获取更多的特征表示。实验表明,本文方法不仅可以获得较高精度,而且能够有效减少模型训练时间,同时也验证了进行域间特征迁移可以加速网络学习与优化这一结论。     

基于环形特征的卷积神经网络轮毂识别

针对实际生产中不同种类轮毂的混流生产问题,提出了一种基于环形特征的卷积神经网络轮毂识别算法。将直角坐标下的环形轮毂映射到极坐标中,归一化为标准形式的矩形,提取轮毂图像的环形特征信息,减少冗余特征产生的影响;设计了一种改进的VGG网络架构,利用深度可分离卷积打破输出通道维度与卷积核大小的联系,在不损失网络性能的同时降低了计算量,能够在实际生产中轮毂识别任务在有限的算力情况下实时进行计算;从有效性和实时性两个方面对轮毂识别算法进行评估,且通过Inception V3、SVM、KNN等模型的对比实验,验证了该算法可以实时地对轮毂自适应分类。实验表明: 该方法对轮毂图像的处理精度达到99%以上,单幅图像平均处理时间降低至11.78ms。     

基于机器视觉的轮毂在线识别分类技术研究

为了对生产线上的轮毂进行识别分类,本文开发了一套基于OpenCV和MFC平台的轮毂型号在线识别系统.首先提取轮毂的高度、外直径、中心孔直径、辐条数目、幅窗的周长面积比等特征参数.其中,通过图像预处理、边缘检测、圆拟合、系统标定等方法获取轮毂外直径,来表征各类轮毂的尺寸;通过提取辐条数目、中心孔直径、幅窗的周长面积比等具有旋转不变性的常量来表征各类轮毂的形状.然后为提取到的特征参数生成序列号,作为型号识别的特征参数.最后将生成的特征序列号与模板库中的标准数值进行比对,达到在线实时分类的效果.实验结果表明:该系统的识别准确率为98.7%,能够有效地完成轮毂的在线识别分类,为轮毂缺陷检测的自动化、智能化提供了保障.     

基于迁移学习的钢金相组织分类与识别方法的研究

金相检验是分析钢内部组织的常用方法,其中检验图像由人工判别,容易受到主观因素的影响而造成结果的不确定.近年来,深度学习(Deep learning,DL)方法中的卷积神经网络(Convolutional neural networks,CNN)能从原始图像中学习复杂的特征,在图像分类与识别领域得到了广泛的应用.CNN建模需要大量的训练样本才能达到较好的泛化能力,材料科学与工程领域针对具体问题的数据集往往较小,不能满足CNN建模的条件,制约了DL在材料领域的应用.本研究基于lmageNet数据集预训练VGG19模型,对火力发电机组耐热钢金相检验图像进行识别,采用冻结全部卷积层权值和微调部分卷积层权值两种迁移学习方法,可以克服金相图像数据集较小的问题,实现小样本数据集的深度学习建模,两种方法的准确率分别为92.5％和94.2％.微调方式的迁移学习CNN模型具有较快的收敛速度、较高的训练精度与泛化能力,能够对金相组织图像进行较为准确的分类与识别,是一种智能的钢金相组织识别方法,也是自动化分类与识别钢金相组织的一种新方法.     

模型无关元迁移学习用于空间滚动轴承寿命阶段识别

针对变工况下空间滚动轴承寿命阶段识别时因样本分布差异较大、可训练用寿命阶段样本较少以及不同寿命阶段样本数量不均等所造成的寿命阶段识别准确率较低的问题,提出模型无关元迁移学习（Model-Agnostic Meta-Transfer Learning, MAMTL）用于空间滚动轴承寿命阶段识别。在MAMTL中,将模型无关元学习和迁移学习相结合以实现多任务同步平行训练从而代替传统的迭代训练,多个任务损失函数利用不同工况下无类标签样本和历史工况下少量有类标签样本共同更新MAMTL网络参数,以寻求网络参数的全局最优解,这使MAMTL具有更好的泛化能力,因此MAMTL在较少历史工况有类标签训练样本情况下比传统迁移学习具有更好的域适配性;在MAMTL中构建新型原型网络以将历史工况每一类别的样本表示为一个原型,通过计算当前工况待测样本与原型的相似度完成当前工况待测样本分类,且该分类过程无需参数学习,因此可防止样本不均等情况下对于不同类别样本识别精度差距较大和在少量有类标签训练样本情况下网络出现过拟合的问题,从而更好提高分类精度。MAMTL的以上优势使得它可利用空间滚动轴承历史工况下的少量、非均等已知...     

基于迁移学习的多模态AD病程分类研究

近年来,患阿尔茨海默病(Alzheimer’s Disease, AD)的人数逐年增加。临床研究显示,轻度认知障碍(Mild Cognitive Impairment, MCI)转化为AD的概率很大,因此,提高磁共振成像(Magnetic Resonance Imaging, MRI)和正电子发射断层扫描(Positron Emission Tomography, PET)等神经影像图对AD、 MCI的分类准确率十分必要。为了解决数据量少、标注困难的问题,首先使用CycleGAN网络对缺少的PET图进行生成;然后采用基于区域能量融合准则的小波变换算法对MRI图和PET图进行图像融合,能够极大程度的保留图像中的数据信息;最后利用迁移学习技术对轻量级网络MobileNet进行训练与微调。实验结果显示,在数据量较少的情况下,所提方法在泛化能力较好的同时,也获得了较高的准确率。     

基于Efficientnet的宫颈图像的识别分类

宫颈癌是全球第二高发的女性癌症,但是如果及时发现,其治愈率几乎为100％.阴道镜检查是临床筛查宫颈上皮内瘤变(CIN)和早期宫颈癌的重要步骤之一,直接影响患者的诊断方案.然而,这种方法取决于阴道镜检查者的观察.本文建立了宫颈图像的数据集,并提出了一种基于Efficientnet的宫颈图像分类的方法.实验结果表明,该模型取得了比经典深度学习方法更好的分类性能,其分类结果准确率可达90.56％.     

基于迁移学习的异步电机故障诊断

针对异步电机故障诊断中,故障数据样本少导致传统深度神经网络模型泛化能力差的问题,提出一种异构迁移学习的异步电机故障诊断算法。首先,通过仿真平台模拟异步电机故障,以解决故障数据样本少的问题;其次,对正常和故障状态下的电流电压信号进行小波变换,作为深度学习网络的输入;然后,基于多核最大平均差异方法,获得仿真数据和实测数据的深度特征差异,对深度学习神经网络参数微调,使其深度学习特征具有跨域不变性。最终,在实验平台上验证文中所提算法,实验结果表明,该算法的故障诊断准确率高,依赖实测故障数据样本少。     

基于改进迁移学习的运动想象分类识别算法北大核心CSCD

利用迁移学习算法提高分类识别的准确率是运动想象脑机接口应用的热点研究问题,其中样本迁移和特征迁移的传统模型算法在样本量较少或源域数据和目标域数据差异较大情况时,各自的迁移效果并不理想。基于欧式对齐(EA)和改进联合类质心匹配和局部流形自学习(CMMS)迁移学习的运动想象分类算法,将样本迁移和特征迁移的优势有机结合,在考虑样本本身的同时,进一步提高了分类准确率。首先,对样本进行源域和目标域的EA,减少源域和目标域的数据分布差异;其次,基于最小化最大均值差异(MMD)改进CMMS方法,筛选源域数据,再次减小源域样本与目标域的分布差异;最后,将该方法应用于BCI竞赛数据集进行离线测试和在线实验。实验结果表明:所研究的方法与SVM、JDA、BDA、EasyTL、GFK、CMMS相比较,迁移学习模型的识别准确率分别提高了14.38%,8.5%,5.8%,10.4%,11.8%,5.7%。     

基于改进YOLOv4算法的轮毂表面缺陷检测

汽车轮毂加工过程中产生的表面缺陷严重影响整车的美观性及服役性能,针对人工检测效率低、漏检率高的问题,提出一种基于改进YOLOv4算法的轮毂表面缺陷检测方法。构建了轮毂缺陷数据集,其包含6种表面缺陷,由2346张4928×3264pixel的图像组成;采用K-means方法进行先验框聚类,并针对YOLOv4算法在纤维、粘铝等小尺度缺陷上检测精度不足问题,在原网络Neck部分引入细化U型网络模块(TUM)和注意力机制,用于增强有效特征并抑制无效特征,强化多尺度特征提取与融合,改善特征处理过程中可能存在的小目标信息丢失问题;基于该数据集,训练并测试不同算法的缺陷检测性能并验证改进模块的有效性。结果表明,该方法大幅提升了粘铝等小尺寸缺陷的检测能力,缺陷检测平均精度达到85.8%,与多种算法相比较检测精度最高。     

轮毂造型设计对自主品牌SUV外观设计的影响探析

目的从轮彀设计层面,为自主品牌SUV提供提升产品价值与认可度的方法。方法采用感性分析法,对当前具有代表性的8款自主品牌SUV轮彀特征进行分析,提出轮彀设计对自主品牌SUV外观整体提升的影响以及解决的方向。结论轮彀设计对自主品牌SUV外观影响主要体现在轮彀尺寸、轮彀造型形式与轮彀色彩搭配。     

Human Gait Recognition Using Deep Learning and Improved Ant Colony Optimization

Human gait recognition (HGR) has received a lot of attention in the last decade as an alternative biometric technique. The main challenges in gait recognition are the change in in-person view angle and covariant factors. The major covariant factors are walking while carrying a bag and walking while wearing a coat. Deep learning is a new machine learning technique that is gaining popularity. Many techniques for HGR based on deep learning are presented in the literature. The requirement of an efficient framework is always required for correct and quick gait recognition. We proposed a fully automated deep learning and improved ant colony optimization (IACO) framework for HGR using video sequences in this work. The proposed framework consists of four primary steps. In the first step, the database is normalized in a video frame. In the second step, two pre-trained models named ResNet101 and InceptionV3 are selected and modified according to the dataset's nature. After that, we trained both modified models using transfer learning and extracted the features. The IACO algorithm is used to improve the extracted features. IACO is used to select the best features, which are then passed to the Cubic SVM for final classification. The cubic SVM employs a multiclass method. The experiment was carried out on three angles (0, 18, and 180) of the CASIA B dataset, and the accuracy was 95.2, 93.9, and 98.2 percent, respectively. A comparison with existing techniques is also performed, and the proposed method outperforms in terms of accuracy and computational time.     

Leveraging Transfer Learning for Spatio-Temporal Human Activity Recognition from Video Sequences

Human Activity Recognition (HAR) is an active research area due to its applications in pervasive computing, human-computer interaction, artificial intelligence, health care, and social sciences. Moreover, dynamic environments and anthropometric differences between individuals make it harder to recognize actions. This study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world applications. It uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural network. Moreover, the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification information. Six state-of-the-art pre-trained models are exploited to find the best model for spatial feature extraction. For temporal sequence, this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short Term Memory (BiLSTM) to capture long-term dependencies. Two state-of-the-art datasets, UCF101 and HMDB51, are used for evaluation purposes. In addition, seven state-of-the-art optimizers are used to fine-tune the proposed network parameters. Furthermore, this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network (CNN), where two streams use RGB data. In contrast, the other uses optical flow images. Finally, the proposed ensemble approach using max hard voting outperforms state-of-the-art methods with 96.30% and 90.07% accuracies on the UCF101 and HMDB51 datasets.     

改进的Inceptionv3果蔬识别算法

目的为了提高果蔬农产品识别的准确性,使果蔬农产品分类实现自动化。方法利用深度卷积神经网路强大的特征学习和特征表达能力,来自动学习果蔬种类特征,提出基于位置的柔性注意力算法,对Inceptionv3神经网络进行改进,并结合参数迁移学习方法建立果蔬识别模型;针对果蔬种类繁多,且国内外缺乏完善的果蔬图像数据库这一现状,构建果蔬图像数据集;在此数据集上将文中所提出的果蔬识别算法与其他果蔬识别算法进行对比。结果试验结果表明,在学习率为0.1、迭代次数为5000时,文中提出算法的准确率高达97.89%。结论相较于现有果蔬识别算法,所提出的果蔬识别算法的识别性能最优,鲁棒性最强。     

Multi-View & Transfer Learning for Epilepsy Recognition Based on EEG Signals

Epilepsy is a central nervous system disorder in which brain activity becomes abnormal. Electroencephalogram (EEG) signals, as recordings of brain activity, have been widely used for epilepsy recognition. To study epileptic EEG signals and develop artificial intelligence (AI)-assist recognition, a multi-view transfer learning (MVTL-LSR) algorithm based on least squares regression is proposed in this study. Compared with most existing multi-view transfer learning algorithms, MVTL-LSR has two merits: (1) Since traditional transfer learning algorithms leverage knowledge from different sources, which poses a significant risk to data privacy. Therefore, we develop a knowledge transfer mechanism that can protect the security of source domain data while guaranteeing performance. (2) When utilizing multi-view data, we embed view weighting and manifold regularization into the transfer framework to measure the views’ strengths and weaknesses and improve generalization ability. In the experimental studies, 12 different simulated multi-view & transfer scenarios are constructed from epileptic EEG signals licensed and provided by the University of Bonn, Germany. Extensive experimental results show that MVTL-LSR outperforms baselines. The source code will be available on .     

基于GA-BP神经网络晶粒尺寸预测模型的轮端轮毂锻造工艺优化

目的 针对6082铝合金轮端轮毂在热处理过程中出现的粗晶问题,利用基于遗传算法优化的BP神经网络晶粒尺寸预测模型模拟优化锻造工艺方案,避免产生粗晶。方法 以遗传算法替代梯度下降法优化神经网络各节点的权值和阈值,建立高精度的GA-BP神经网络晶粒尺寸预测模型,再以轮端轮毂为对象,设计锻造工艺方案并利用Deform进行微观组织仿真,研究压下速率、坯料初始温度对晶粒尺寸的影响,获得最优方案。结果 优化模型预测的晶粒尺寸平均值和最大值的平均绝对百分比误差分别为2.55%、0.43%,与常规的BP神经网络相比,准确性有了较大提高。对比不同锻造方案的结果,得到轮毂较优的初始坯料温度为500℃,压下速率为200 mm/s,经试验验证,锻件特征位置的晶粒尺寸预测值与实际值之间的误差均在10%以下,表明该预测模型具有良好的工程应用价值。结论 遗传算法的引入大大增强了BP神经网络的全局寻优能力,提高了模型的准确性。在Deform中复现的预测模型对锻件的晶粒尺寸分布有较好的预测效果,并基于此成功模拟、优化了轮端轮毂的锻造方案。     

Image Recognition of Citrus Diseases Based on Deep Learning

In recent years, with the development of machine learning and deep learning, it is possible to identify and even control crop diseases by using electronic devices instead of manual observation. In this paper, an image recognition method of citrus diseases based on deep learning is proposed. We built a citrus image dataset including six common citrus diseases. The deep learning network is used to train and learn these images, which can effectively identify and classify crop diseases. In the experiment, we use MobileNetV2 model as the primary network and compare it with other network models in the aspect of speed, model size, accuracy. Results show that our method reduces the prediction time consumption and model size while keeping a good classification accuracy. Finally, we discuss the significance of using MobileNetV2 to identify and classify agricultural diseases in mobile terminal, and put forward relevant suggestions.     

Multi-Label Learning Based on Transfer Learning and Label Correlation

In recent years, multi-label learning has received a lot of attention. However, most of the existing methods only consider global label correlation or local label correlation. In fact, on the one hand, both global and local label correlations can appear in real-world situation at same time. On the other hand, we should not be limited to pairwise labels while ignoring the high-order label correlation. In this paper, we propose a novel and effective method called GLLCBN for multi-label learning. Firstly, we obtain the global label correlation by exploiting label semantic similarity. Then, we analyze the pairwise labels in the label space of the data set to acquire the local correlation. Next, we build the original version of the label dependency model by global and local label correlations. After that, we use graph theory, probability theory and Bayesian networks to eliminate redundant dependency structure in the initial version model, so as to get the optimal label dependent model. Finally, we obtain the feature extraction model by adjusting the Inception V3 model of convolution neural network and combine it with the GLLCBN model to achieve the multi-label learning. The experimental results show that our proposed model has better performance than other multi-label learning methods in performance evaluating.