基于深度学习的情感化设计

目的使用深度学习实现情感化设计,满足用户个性化的情感需求,加速传统设计过程,提升用户体验。方法研究深度学习中可用于情感化设计的算法,使用卷积神经网络(CNN)实现名画复制品的个性化自动生成;抓取互联网数据,使用LSTM模型挖掘用户真实需求,进行当前流行游戏的周边产品设计;以张裕葡萄酒庄旅游纪念品设计为例,使用深度学习基于用户个人信息和行为数据推荐个性化的葡萄酒包装。结论基于CNN的名画复制品的个性化生成丰富了图像的可修改空间,满足了用户个性化的情感诉求;基于LSTM的用户需求分析高效和准确地反映了用户的真实需求,加速了传统用户调研过程;基于深度学习的旅游纪念品个性化设计进一步提升了用户体验。将深度学习应用于情感化设计有利于挖掘用户内心的真实需求,节省人力物力,满足用户情感诉求和提升用户体验,进一步为设计学与计算机科学的交叉提供了有效方法。     

Electroencephalogram (EEG) Brain Signals to Detect Alcoholism Based on Deep Learning

The detection of alcoholism is of great importance due to its effects on individuals and society. Automatic alcoholism detection system (AADS) based on electroencephalogram (EEG) signals is effective, but the design of a robust AADS is a challenging problem. AADS’ current designs are based on conventional, hand-engineered methods and restricted performance. Driven by the excellent deep learning (DL) success in many recognition tasks, we implement an AAD system based on EEG signals using DL. A DL model requires huge number of learnable parameters and also needs a large dataset of EEG signals for training which is not easy to obtain for the AAD problem. In order to solve this problem, we propose a multi-channel Pyramidal neural convolutional (MP-CNN) network that requires a less number of learnable parameters. Using the deep CNN model, we build an AAD system to detect from EEG signal segments whether the subject is alcoholic or normal. We validate the robustness and effectiveness of proposed AADS using KDD, a benchmark dataset for alcoholism detection problem. In order to find the brain region that contributes significant role in AAD, we investigated the effects of selected 19 EEG channels (SC-19), those from the whole brain (ALL-61), and 05 brain regions, i.e., TEMP, OCCIP, CENT, FRONT, and PERI. The results show that SC-19 contributes significant role in AAD with the accuracy of 100%. The comparison reveals that the state-of-the-art systems are outperformed by the AADS. The proposed AADS will be useful in medical diagnosis research and health care systems.     

Automated Chinese Essay Scoring Based on Deep Learning

Writing is an important part of language learning and is considered the best approach to demonstrate the comprehensive language skills of students. Manually grading student essays is a time-consuming task; however, it is necessary. An automated essay scoring system can not only greatly improve the efficiency of essay scoring, but also provide more objective score. Therefore, many researchers have been exploring automated essay scoring techniques and tools. However, the technique of scoring Chinese essays is still limited, and its accuracy needs to be enhanced further. To improve the accuracy of the scoring model for a Chinese essay, we propose an automated scoring approach based on a deep learning model and validate its effect by conducting two comparison experiments. The experimental results indicate that the accuracy of the proposed model is significantly higher than that of multiple linear regression (MLR), which was commonly used in the past. The three accuracy rates of the proposed model are comparable to those of the novice teacher. The root mean square error (RMSE) of the proposed model is slightly lower than that of the novice teacher, and the correlation coefficient of the proposed model is also significantly higher than that of the novice teacher. Besides, when the predicted scores are not very low or very high, the two predicted models are as good as a novice teacher. However, when the predicted score is very high or very low, the results should be treated with caution.     

Autonomous Parking-Lots Detection with Multi-Sensor Data Fusion Using Machine Deep Learning Techniques

The rapid development and progress in deep machine-learning techniques have become a key factor in solving the future challenges of humanity. Vision-based target detection and object classification have been improved due to the development of deep learning algorithms. Data fusion in autonomous driving is a fact and a prerequisite task of data preprocessing from multi-sensors that provide a precise, well-engineered, and complete detection of objects, scene or events. The target of the current study is to develop an in-vehicle information system to prevent or at least mitigate traffic issues related to parking detection and traffic congestion detection. In this study we examined to solve these problems described by (1) extracting region-of-interest in the images (2) vehicle detection based on instance segmentation, and (3) building deep learning model based on the key features obtained from input parking images. We build a deep machine learning algorithm that enables collecting real video-camera feeds from vision sensors and predicting free parking spaces. Image augmentation techniques were performed using edge detection, cropping, refined by rotating, thresholding, resizing, or color augment to predict the region of bounding boxes. A deep convolutional neural network F-MTCNN model is proposed that simultaneously capable for compiling, training, validating and testing on parking video frames through video-camera. The results of proposed model employing on publicly available PK-Lot parking dataset and the optimized model achieved a relatively higher accuracy 97.6% than previous reported methodologies. Moreover, this article presents mathematical and simulation results using state-of-the-art deep learning technologies for smart parking space detection. The results are verified using Python, TensorFlow, OpenCV computer simulation frameworks.     

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。     

Intelligent Decision Support System for COVID-19 Empowered with Deep Learning

The prompt spread of Coronavirus (COVID-19) subsequently adorns a big threat to the people around the globe. The evolving and the perpetually diagnosis of coronavirus has become a critical challenge for the healthcare sector. Drastically increase of COVID-19 has rendered the necessity to detect the people who are more likely to get infected. Lately, the testing kits for COVID-19 are not available to deal it with required proficiency, along with-it countries have been widely hit by the COVID-19 disruption. To keep in view the need of hour asks for an automatic diagnosis system for early detection of COVID-19. It would be a feather in the cap if the early diagnosis of COVID-19 could reveal that how it has been affecting the masses immensely. According to the apparent clinical research, it has unleashed that most of the COVID-19 cases are more likely to fall for a lung infection. The abrupt changes do require a solution so the technology is out there to pace up, Chest X-ray and Computer tomography (CT) scan images could significantly identify the preliminaries of COVID-19 like lungs infection. CT scan and X-ray images could flourish the cause of detecting at an early stage and it has proved to be helpful to radiologists and the medical practitioners. The unbearable circumstances compel us to flatten the curve of the sufferers so a need to develop is obvious, a quick and highly responsive automatic system based on Artificial Intelligence (AI) is always there to aid against the masses to be prone to COVID-19. The proposed Intelligent decision support system for COVID-19 empowered with deep learning (ID2S-COVID19-DL) study suggests Deep learning (DL) based Convolutional neural network (CNN) approaches for effective and accurate detection to the maximum extent it could be, detection of coronavirus is assisted by using X-ray and CT-scan images. The primary experimental results here have depicted the maximum accuracy for training and is around 98.11 percent and for validation it comes out to be approximately 95.5 percent while statistical parameters like sensitivity and specificity for training is 98.03 percent and 98.20 percent respectively, and for validation 94.38 percent and 97.06 percent respectively. The suggested Deep Learning-based CNN model unleashed here opts for a comparable performance with medical experts and it is helpful to enhance the working productivity of radiologists. It could take the curve down with the downright contribution of radiologists, rapid detection of COVID-19, and to overcome this current pandemic with the proven efficacy.     

Image-Based Automatic Diagnostic System for Tomato Plants Using Deep Learning

Tomato production is affected by various threats, including pests, pathogens, and nutritional deficiencies during its growth process. If control is not timely, these threats affect the plant-growth, fruit-yield, or even loss of the entire crop, which is a key danger to farmers’ livelihood and food security. Traditional plant disease diagnosis methods heavily rely on plant pathologists that incur high processing time and huge cost. Rapid and cost-effective methods are essential for timely detection and early intervention of basic food threats to ensure food security and reduce substantial economic loss. Recent developments in Artificial Intelligence (AI) and computer vision allow researchers to develop image-based automatic diagnostic tools to quickly and accurately detect diseases. In this work, we proposed an AI-based approach to detect diseases in tomato plants. Our goal is to develop an end-to-end system to diagnose essential crop problems in real-time, ensuring high accuracy. This paper employs various deep learning models to recognize and predict different diseases caused by pathogens, pests, and nutritional deficiencies. Various Convolutional Neural Networks (CNNs) are trained on a large dataset of leaves and fruits images of tomato plants. We compared the performance of ShallowNet (a shallow network trained from scratch) and the state-of-the-art deep learning network (models are fine-tuned via transfer learning). In our experiments, DenseNet consistently achieved high performance with an accuracy score of 95.31% on the test dataset. The results verify that deep learning models with the least number of parameters, reasonable complexity, and appropriate depth achieve the best performance. All experiments are implemented in Python, utilizing the Keras deep learning library backend with TensorFlow.     

Deep Learning ResNet101 Deep Features of Portable Chest X-Ray Accurately Classify COVID-19 Lung Infection

This study is designed to develop Artificial Intelligence (AI) based analysis tool that could accurately detect COVID-19 lung infections based on portable chest x-rays (CXRs). The frontline physicians and radiologists suffer from grand challenges for COVID-19 pandemic due to the suboptimal image quality and the large volume of CXRs. In this study, AI-based analysis tools were developed that can precisely classify COVID-19 lung infection. Publicly available datasets of COVID-19 (N = 1525), non-COVID-19 normal (N = 1525), viral pneumonia (N = 1342) and bacterial pneumonia (N = 2521) from the Italian Society of Medical and Interventional Radiology (SIRM), Radiopaedia, The Cancer Imaging Archive (TCIA) and Kaggle repositories were taken. A multi-approach utilizing deep learning ResNet101 with and without hyperparameters optimization was employed. Additionally, the features extracted from the average pooling layer of ResNet101 were used as input to machine learning (ML) algorithms, which twice trained the learning algorithms. The ResNet101 with optimized parameters yielded improved performance to default parameters. The extracted features from ResNet101 are fed to the k-nearest neighbor (KNN) and support vector machine (SVM) yielded the highest 3-class classification performance of 99.86% and 99.46%, respectively. The results indicate that the proposed approach can be better utilized for improving the accuracy and diagnostic efficiency of CXRs. The proposed deep learning model has the potential to improve further the efficiency of the healthcare systems for proper diagnosis and prognosis of COVID-19 lung infection.     

基于深度学习的界面设计美学评价研究

目的美感已经成为人机交互(HCI)的核心结构之一,对用户的感知和态度具有明显的有益影响。然而界面美观性评价方法仍是设计师及其团队所面临的重要问题。引入深度学习技术来探讨其评价界面设计美感的可能性。方法分别使用基于深度卷积神经网络的闪屏美学分类方法和Google提出的基于深度学习NIMA神经网络,来预测闪屏图像的美学评价分布。结果通过研究发现,使用基于深度学习NIMA神经网络可以得到比传统方法更具体的评价结果,帮助设计师有效而客观地评价界面设计。结论将计算机图像美学评价的研究领域拓展到界面设计领域,验证了深度卷积神经网络在界面设计美学评价领域使用的可行性。未来图像美学评价还可以介入更多的设计相关领域,辅助设计师做出更有效的设计和商业决策。     

Deep Learning Based Face Detection and Identification of Criminal Suspects

Occurrence of crimes has been on the constant rise despite the emerging discoveries and advancements in the technological field in the past decade. One of the most tedious tasks is to track a suspect once a crime is committed. As most of the crimes are committed by individuals who have a history of felonies, it is essential for a monitoring system that does not just detect the person’s face who has committed the crime, but also their identity. Hence, a smart criminal detection and identification system that makes use of the OpenCV Deep Neural Network (DNN) model which employs a Single Shot Multibox Detector for detection of face and an auto-encoder model in which the encoder part is used for matching the captured facial images with the criminals has been proposed. After detection and extraction of the face in the image by face cropping, the captured face is then compared with the images in the Criminal Database. The comparison is performed by calculating the similarity value between each pair of images that are obtained by using the Cosine Similarity metric. After plotting the values in a graph to find the threshold value, we conclude that the confidence rate of the encoder model is 0.75 and above.     

Real-Time Violent Action Recognition Using Key Frames Extraction and Deep Learning

Violence recognition is crucial because of its applications in activities related to security and law enforcement. Existing semi-automated systems have issues such as tedious manual surveillances, which causes human errors and makes these systems less effective. Several approaches have been proposed using trajectory-based, non-object-centric, and deep-learning-based methods. Previous studies have shown that deep learning techniques attain higher accuracy and lower error rates than those of other methods. However, the their performance must be improved. This study explores the state-of-the-art deep learning architecture of convolutional neural networks (CNNs) and inception V4 to detect and recognize violence using video data. In the proposed framework, the keyframe extraction technique eliminates duplicate consecutive frames. This keyframing phase reduces the training data size and hence decreases the computational cost by avoiding duplicate frames. For feature selection and classification tasks, the applied sequential CNN uses one kernel size, whereas the inception v4 CNN uses multiple kernels for different layers of the architecture. For empirical analysis, four widely used standard datasets are used with diverse activities. The results confirm that the proposed approach attains 98% accuracy, reduces the computational cost, and outperforms the existing techniques of violence detection and recognition.     

Tissue Segmentation in Nasopharyngeal CT Images Using TwoStage Learning

Tissue segmentation is a fundamental and important task in nasopharyngeal images analysis. However, it is a challenging task to accurately and quickly segment various tissues in the nasopharynx region due to the small difference in gray value between tissues in the nasopharyngeal image and the complexity of the tissue structure. In this paper, we propose a novel tissue segmentation approach based on a two-stage learning framework and U-Net. In the proposed methodology, the network consists of two segmentation modules. The first module performs rough segmentation and the second module performs accurate segmentation. Considering the training time and the limitation of computing resources, the structure of the second module is simpler and the number of network layers is less. In addition, our segmentation module is based on U-Net and incorporates a skip structure, which can make full use of the original features of the data and avoid feature loss. We evaluated our proposed method on the nasopharyngeal dataset provided by West China Hospital of Sichuan University. The experimental results show that the proposed method is superior to many standard segmentation structures and the recently proposed nasopharyngeal tissue segmentation method, and can be easily generalized across different tissue types in various organs.     

基于视觉感知的机器人工件识别方法研究

目的 解决定制化木门尺寸规格不统一、表面纹理多样而导致的堆垛分类困难、搬运效率低下等问题。方法 提出采用深度学习方法进行定制式木门工件检测,以YOLOV3网络为基本框架开展机器人工件识别方法研究。首先,通过图像数据增强和预处理,扩充定制式木门数据;然后,进行YOLO V3损失函数改进,并根据木门特征进行定制式木门数据集锚框尺度的重新聚类;最后,应用空间金字塔池化层进行YOLO V3中特征金字塔网络改进,并通过随机选取的测试集验证本文方法的有效性。结果 测试数据集的平均检测准确率均值达到98.05%,检测每张图片的时间为137 ms。结论 研究表明,本文方法能够满足木门生产线对准确率和实时性的要求,可大大提高定制化木门转线及堆垛效率。     

Deep convolutional neural networks for eigenvalue problems in mechanics

We show that deep convolutional neural networks (CNNs) can massively outperform traditional densely connected neural networks (NNs) (both deep or shallow) in predicting eigenvalue problems in mechanics. In this sense, we strike out in a new direction in mechanics computations with strongly predictive NNs whose success depends not only on architectures being deep but also being fundamentally different from the widely used to date. We consider a model problem: predicting the eigenvalues of one-dimensional (1D) and two-dimensional (2D) phononic crystals. For the 1D case, the optimal CNN architecture reaches 98% accuracy level on unseen data when trained with just 20 000 samples, compared to 85% accuracy even with 100 000 samples for the typical network of choice in mechanics research. We show that, with relatively high data efficiency, CNNs have the capability to generalize well and automatically learn deep symmetry operations, easily extending to higher dimensions and our 2D case. Most importantly, we show how CNNs can naturally represent mechanical material tensors, with its convolution kernels serving as local receptive fields, which is a natural representation of mechanical response. Strategies proposed are applicable to other mechanics' problems and may, in the future, be used to sidestep cumbersome algorithms with purely data-driven approaches based upon modern deep architectures.     

Software Defect Prediction Based on Non-Linear Manifold Learning and Hybrid Deep Learning Techniques

Software defect prediction plays a very important role in software quality assurance, which aims to inspect as many potentially defect-prone software modules as possible. However, the performance of the prediction model is susceptible to high dimensionality of the dataset that contains irrelevant and redundant features. In addition, software metrics for software defect prediction are almost entirely traditional features compared to the deep semantic feature representation from deep learning techniques. To address these two issues, we propose the following two solutions in this paper: (1) We leverage a novel non-linear manifold learning method - SOINN Landmark Isomap (SLIsomap) to extract the representative features by selecting automatically the reasonable number and position of landmarks, which can reveal the complex intrinsic structure hidden behind the defect data. (2) We propose a novel defect prediction model named DLDD based on hybrid deep learning techniques, which leverages denoising autoencoder to learn true input features that are not contaminated by noise, and utilizes deep neural network to learn the abstract deep semantic features. We combine the squared error loss function of denoising autoencoder with the cross entropy loss function of deep neural network to achieve the best prediction performance by adjusting a hyperparameter. We compare the SL-Isomap with seven state-of-the-art feature extraction methods and compare the DLDD model with six baseline models across 20 open source software projects. The experimental results verify that the superiority of SL-Isomap and DLDD on four evaluation indicators.     

Deep Feature Extraction and Feature Fusion for Bi-Temporal Satellite Image Classification

Multispectral images contain a large amount of spatial and spectral data which are effective in identifying change areas. Deep feature extraction is important for multispectral image classification and is evolving as an interesting research area in change detection. However, many deep learning framework based approaches do not consider both spatial and textural details into account. In order to handle this issue, a Convolutional Neural Network (CNN) based multi-feature extraction and fusion is introduced which considers both spatial and textural features. This method uses CNN to extract the spatio-spectral features from individual channels and fuse them with the textural features. Then the fused image is classified into change and unchanged regions. The presence of mixed pixels in the bitemporal satellite images affect the classification accuracy due to the misclassification errors. The proposed method was compared with six state-of-theart change detection methods and analyzed. The main highlight of this method is that by taking into account the spatio-spectral and textural information in the input channels, the mixed pixel problem is solved. Experiments indicate the effectiveness of this method and demonstrate that it possesses low misclassification errors, higher overall accuracy and kappa coefficient.     

Power System Resiliency and Wide Area Control Employing Deep Learning Algorithm

The power transfer capability of the smart transmission grid-connected networks needs to be reduced by inter-area oscillations. Due to the fact that inter-area modes of oscillations detain and make instability of power transmission networks. This fact is more noticeable in smart grid-connected systems. The smart grid infrastructure has more renewable energy resources installed for its operation. To overcome this problem, a deep learning wide-area controller is proposed for real-time parameter control and smart power grid resilience on oscillations inter-area modes. The proposed Deep Wide Area Controller (DWAC) uses the Deep Belief Network (DBN). The network weights are updated based on real-time data from Phasor measurement units. Resilience assessment based on failure probability, financial impact, and time-series data in grid failure management determine the norm H₂. To demonstrate the effectiveness of the proposed framework, a time-domain simulation case study based on the IEEE-39 bus system was performed. For a one-channel attack on the test system, the resiliency index increased to 0.962, and inter-area damping ξ was reduced to 0.005. The obtained results validate the proposed deep learning algorithm’s efficiency on damping inter-area and local oscillation on the 2-channel attack as well. Results also offer robust management of power system resilience and timely control of the operating conditions.     

Feature Fusion-Based Deep Learning Network to Recognize Table Tennis Actions

A system for classifying four basic table tennis strokes using wearable devices and deep learning networks is proposed in this study. The wearable device consisted of a six-axis sensor, Raspberry Pi 3, and a power bank. Multiple kernel sizes were used in convolutional neural network (CNN) to evaluate their performance for extracting features. Moreover, a multiscale CNN with two kernel sizes was used to perform feature fusion at different scales in a concatenated manner. The CNN achieved recognition of the four table tennis strokes. Experimental data were obtained from 20 research participants who wore sensors on the back of their hands while performing the four table tennis strokes in a laboratory environment. The data were collected to verify the performance of the proposed models for wearable devices. Finally, the sensor and multi-scale CNN designed in this study achieved accuracy and F1 scores of 99.58% and 99.16%, respectively, for the four strokes. The accuracy for five-fold cross validation was 99.87%. This result also shows that the multi-scale convolutional neural network has better robustness after five-fold cross validation.     

GDX1包装机IPC控制系统的研发

目的 为加强银行智能办理业务的设备性能,提高票据数字的识别效率,研究一种改进的算法来获得更高的数字识别效果。方法 根据银行票据的印刷数字特性进行字符的提取和分割,经过图像采集、降噪、二值化之后使用起点直方图法结合步长法进行字符的分割,然后使用改进的LENET卷积神经网络用于提取数字特征,进行分类。结果 通过实验,结果表明文中提出的方法进行复杂环境下的印刷数字识别,准确率达到95%以上,识别速率为1.169 s/张。结论 利用新的字符分割算法与改进的LENET神经网络相结合,可以很好地识别干扰强的印刷票据,准确率高。