基于深度学习的排水管道缺陷内窥检测智能识别系统研究

目前国内各城市已普遍采用管道机器人深入管道内部摄取视频影像,有效获取到可供管道缺陷检测的一手资料,但缺陷识别大部分依靠人工目视识别,耗时耗力,生产周期长。利用福州市勘测院多年累积的管道检测数据,基于Pytorch深度学习框架、建立了排水管道缺陷内窥检测智能识别系统,包括:数据预处理,残差神经网络设计与训练、系统集成等。重点实现了三级组合识别模型建构(二分类,类型识别,等级识别),解决了系统准确度等技术难题。经生产实践表明:模型准确率高,可有效提高管道健康状况检查质量和效率。     

基于深度学习的火焰分割模型对比研究

由于火焰分割数据集欠缺,经典语义分割模型在火焰分割的研究应用面小,模型对比实验不充分.针对这些问题,在构建火焰分割数据集的基础上,选用在公开数据集中表现良好的4种语义分割模型和2种骨干网络进行训练和测试,并在不同的应用场景下进行对比实验及分析.实验结果表明,U-Net模型在火焰分割领域取得了较好的效果,其中U-Net+...     

基于深度学习的大区域遥感影像路网提取方法

针对传统的遥感影像道路网分割方法鲁棒性差且难以挖掘影像中深层特征等问题,提出了一种基于深度学习语义分割的道路网提取方法,以Deeplabv3+语义分割模型为基础,采用双次迁移学习方式进行训练,通过多尺度预测进行结果融合,实现对预测结果的优化。针对训练样本标注工作繁琐,基于公开影像及矢量数据源,设计了大规模样本采集方法。针对大区域范围,设计基于网格划分的逐网格分割方法,实现大范围道路提取。基于瓦片地图思路,设计大规模分割数据存储方法,实现分割数据结果的存储管理。通过采集某区域500km~2样本数据,验证了大规模样本采集方法的有效性。对该区域样本数据集训练分析,单尺度下道路提取精度MIoU达到77. 2%,多尺度融合预测可提升1. 1个百分点。最后,基于设计的道路分割系统,可视化验证了网格划分及大规模分割数据存储方法的有效性。     

低应变检测缺陷深度不确定度分析与应用

低应变检测判断桩基缺陷的误差较大,本文试图从影响判断缺陷深度的测量不确定度的几个分量入手,来分析其测量不确定度,旨在供同行在分析低应变检测缺陷深度的测量不确定度及其应用时参考.     

基于计算机视觉的混凝土缺陷检测研究综述

混凝土缺陷对混凝土结构的安全性和稳定性造成的威胁不容小觑,因此,定期的缺陷检测对混凝土结构的维护至关重要。相较于主观低效的人工视觉检测,计算机视觉因在混凝土缺陷检测的自动化方面具有显著优势而成为近年来的研究热点,但目前缺乏该领域的全面综述。因此,本文旨在综合分析计算机视觉技术在混凝土缺陷检测领域的研究进展,对混凝土缺陷检测涉及的计算机视觉算法进行分类,总结现存的技术难点并分析未来研究方向,为该领域的后续研究提供一定的参考。     

基于交通视频和深度学习的车辆检测和跟踪1

车辆目标检测和跟踪是智能交通的关键技术,目前已有的车辆检测和跟踪算法种类繁多但是性能各异,难以同时满足交通视频监控中的实时性和高精度要求.该文采集多段交通监控视频,标注多种类型的车辆目标.在此数据集上从多种维度考查多种基于深度学习的车辆检测算法和多种流行的跟踪算法在交通视频上的表现,其中SSD算法满足实时性要求且mAP达0.878.并提出基于SSD和MEDIANFLOW的车流量实时检测方法.经实验证明,该方法在保证实时性情况下车流量的检测准确率达到94.5％.     

基于深度学习的疑似违法图斑快速检测研究

无人机监测是当前城市违法建设和违法用地问题治理的重要手段,但传统人工识别的方法费时费力,越来越难以满足治理需求.本文针对这一问题,研究了基于深度学习卷积神经网络模型的城市违法建设和违法用地快速检测方法:首先,分析了违法建设和违法用地的主要类型和特点,构建训练样本集;然后,构建卷积神经网络深度学习模型并对模型进行训练;最后,对两期影像分别分类并通过分类结果对比的方式快速筛查违法建设图斑.利用0.1 m分辨率的城市无人机正射影像进行的检测实验结果表明,本文方法对于违法建设问题能够快速有效地检测,对城市违法建设问题治理具有良好的支持作用.     

基于深度学习的居住建筑外墙面损伤检测研究

以Tensorflow 2.0为平台,通过Faster RCNN算法框架建立深度学习模型.以1620张居住建筑外墙面受损照片为数据集.选取其中1296张为训练集,对模型进行有监督训练并测试模型训练深度,324张为测试集校检模型精度.测试结果表明,深度学习模型对居住建筑外墙的污染类损伤检测率为88.82％;裂缝类损伤检测率为90.21％;破损类损伤检测率为90.94％,检测平均耗时为每图0.23s.深度学习检测模型可有效反馈外墙面的主要损伤情况,提高建筑工程管理效率.     

基于深度学习的VGG16 图像型火灾探测方法研究

为了快速、有效地检测不同场景下的火灾信息,基于深度迁移学习设计了一种改进VGG16 的图像型火灾检测方法。搜集不同场景下的照片,使用离线数据增强技术增加样本数量,对VGG16 进行改进,并使用迁移学习的方法训练火灾识别模型。结果表明：改进的VGG16 网络对于火灾现场的图片分类识别准确率为98.7%,优于Resnet50 网络和Densenet121 网络,可快速、准确地检测到火灾信息。     

基于深度学习的人脸识别研究

基于深度学习的人脸识别技术是目前人工智能和图像领域研究的热点之一,尤其随着近年来深度神经网络的发展,人脸识别的准确性和有效性得到了极大的提高。文章首先简要阐述了人脸识别技术的研究和发展历史,接着叙述了人脸识别的技术流程,随后详细介绍了在人脸识别中常用到的卷积神经网络。由于各大企业在人脸识别领域取得丰硕的研究成果,因此,也对人脸识别的产品和公司进行了简单介绍。最后,对人脸识别技术存在的不足和发展前景进行了总结和展望。     

Digital image correlation-based structural state detection through deep learning

This paper presents a new approach for automatical classification of structural state through deep learning. In this work, a Convolutional Neural Network (CNN) was designed to fuse both the feature extraction and classification blocks into an intelligent and compact learning system and detect the structural state of a steel frame; the input was a series of vibration signals, and the output was a structural state. The digital image correlation (DIC) technology was utilized to collect vibration information of an actual steel frame, and subsequently, the raw signals, without further pre-processing, were directly utilized as the CNN samples. The results show that CNN can achieve 99% classification accuracy for the research model. Besides, compared with the backpropagation neural network (BPNN), the CNN had an accuracy similar to that of the BPNN, but it only consumes 19% of the training time. The outputs of the convolution and pooling layers were visually displayed and discussed as well. It is demonstrated that: 1) the CNN can extract the structural state information from the vibration signals and classify them; 2) the detection and computational performance of the CNN for the incomplete data are better than that of the BPNN; 3) the CNN has better anti-noise ability.     

Dynamics-based cross-domain structural damage detection through deep transfer learning

Structural damage detection (SDD) still suffers from environmental uncertainties or modeling errors, causing a gap between the numerical model and the real structure. It results in performance degradation in the application of many model-based methods, which are usually designed on a numerical model and needed to be applied to a real structure. Such a situation is defined as a cross-domain SDD problem in this work. This paper aims to address the cross-domain SDD problem by designing a feature-extractor to generate both damage-sensitive and domain-invariant features, instead of trying to reduce the gap, as the traditional methods do. A domain adaptation (DA) neural network is designed and trained on the data from both the numerical model and the real structure at the same time. In addition, no damage label of the real structure is needed. Both numerical and laboratory experiments show that the proposed method has excellent performance and outperforms the baseline model, a traditional convolutional neural network (CNN). This paper provides a new methodology to solve the cross-domain SDD problem, that is, to learn better features instead of just trying to reduce the gap.     

基于深度学习的渔业领域命名实体识别

为了解决基于分词的渔业领域命名实体识别效果受分词准确度影响这一问题,采用一种基于深度学习的渔业领域命名实体识别方法。该方法使用神经网络训练得到字向量作为模型输入,避免了分词不准确对渔业领域命名实体识别效果造成的影响;针对渔业领域命名实体长度较长这一特点,使用LSTM单元保持较长时间记忆信息,并将标记信息融入到CRF模型中构建Character+LSTM+CRF实体识别模型。为验证方法的有效性,在渔业领域语料集上进行多组实验,结果表明,本研究中提出的Character+LSTM+CRF方法具有较好的效果,与LSTM模型相比较,在准确率、召回率、F值上分别提升了3.39%、2.99%、3.19%,对于渔业领域实体识别具有较好的效果。     

充气膜建筑的柔性薄膜太阳能供电系统研究

介绍了充气膜建筑和柔性薄膜太阳能发电技术的特点及应用,通过负荷计算、光伏阵列设计、发电量计算、设备选型,设计了一种可行的充气膜建筑能耗解决方案,做到了气膜系统建筑能耗上的自给自足,使得充气膜建筑成为零能耗的纯绿色建筑。     

融合改进梅尔谱特征和深信念网络的语音测谎算法

为实现非接触式谎言检测,特提出了以语谱特征为线索,结合深度学习的谎言检测方法。为提取谎言中微颤抖所引起的语谱局部能量变化,算法先对梅尔频谱进行了Hu矩处理,然后进行离散余弦变换去除相关性。该特征利用了Hu矩的正交不变性和平移不变性,能较好的体现出语谱中局部能量的集中方式。然后将所提取的特征作为改进深信念网络输入进行谎言识别。为提高受限玻尔兹曼机的并行回火训练算法中相邻温度链之间的交换率,训练算法先对Markov链的状态能量进行等能量的划分,使得每个能量环内的状态具有相似的能量,然后再进行交换以提高交换率从而优化整个网络的训练。在Columbia-SRI-Colorado数据库上的实验表明,谎言识别率达到了71.47%,比梅尔倒谱系数特征的识别率提高了3%,比传统的BayesNet分类算法提高了7%。     

Efficient Identification of water conveyance tunnels siltation based on ensemble deep learning

The inspection of water conveyance tunnels plays an important role in water diversion projects. Siltation is an essential factor threatening the safety of water conveyance tunnels. Accurate and efficient identification of such siltation can reduce risks and enhance safety and reliability of these projects. The remotely operated vehicle (ROV) can detect such siltation. However, it needs to improve its intelligent recognition of image data it obtains. This paper introduces the idea of ensemble deep learning. Based on the VGG16 network, a compact convolutional neural network (CNN) is designed as a primary learner, called Silt-net, which is used to identify the siltation images. At the same time, the fully-connected network is applied as the meta-learner, and stacking ensemble learning is combined with the outputs of the primary classifiers to obtain satisfactory classification results. Finally, several evaluation metrics are used to measure the performance of the proposed method. The experimental results on the siltation dataset show that the classification accuracy of the proposed method reaches 97.2%, which is far better than the accuracy of other classifiers. Furthermore, the proposed method can weigh the accuracy and model complexity on a platform with limited computing resources.     

Interpretable deep learning for roof fall hazard detection in underground mines

Roof falls due to geological conditions are major hazards in the mining industry, causing work time loss, injuries, and fatalities. There are roof fall problems caused by high horizontal stress in several large-opening limestone mines in the eastern and midwestern United States. The typical hazard management approach for this type of roof fall hazards relies heavily on visual inspections and expert knowledge. In this context, we proposed a deep learning system for detection of the roof fall hazards caused by high horizontal stress. We used images depicting hazardous and non-hazardous roof conditions to develop a convolutional neural network (CNN) for autonomous detection of hazardous roof conditions. To compensate for limited input data, we utilized a transfer learning approach. In the transfer learning approach, an already-trained network is used as a starting point for classification in a similar domain. Results show that this approach works well for classifying roof conditions as hazardous or safe, achieving a statistical accuracy of 86.4%. This result is also compared with a random forest classifier, and the deep learning approach is more successful at classification of roof conditions. However, accuracy alone is not enough to ensure a reliable hazard management system. System constraints and reliability are improved when the features used by the network are understood. Therefore, we used a deep learning interpretation technique called integrated gradients to identify the important geological features in each image for prediction. The analysis of integrated gradients shows that the system uses the same roof features as the experts do on roof fall hazards detection. The system developed in this paper demonstrates the potential of deep learning in geotechnical hazard management to complement human experts, and likely to become an essential part of autonomous operations in cases where hazard identification heavily depends on expert knowledge. Moreover, deep learning-based systems reduce expert exposure to hazardous conditions.