GAN-based data augmentation of prohibited item X-ray images in security inspection

Convolutional neural networks (CNNs) based methods for automatic discriminant of prohibited items in X-ray images attract attention increasingly. However, it is difficult to train a reliable CNN model using the available X-ray security image databases, since they are not enough in sample quantity and diversity. Recently, generative adversarial network (GAN) has been widely used in image generation and regarded as a power model for data augmentation. In this paper, we propose a data augmentation method for X-ray prohibited item images based on GAN. First, the network structure and loss function of the self-attention generative adversarial network (SAGAN) are improved to generate the realistic X-ray prohibited item images. Then, the images generated by our model are evaluated using GAN-train and GAN-test. Experimental results of GAN-train and GAN-test are 99.91% and 98.82% respectively. It implies that our model can enlarge the X-ray prohibited item image database effectively.     

基于多尺度圆周频率滤波与卷积神经网络的遥感图像飞机目标检测方法研究

针对遥感图像飞机目标检测因目标尺度不一存在漏警、虚警等问题,该文基于遥感图像中飞机目标形状特征和灰度变化特点提出了一种多尺度圆周频率滤波(MSCFF)与卷积神经网络(CNN)相结合的MSCFF+CNN飞机目标自动检测算法.该算法首先采用多尺度圆周频率滤波器滤除遥感图像复杂背景,实现不同尺度飞机目标候选区域提取;然后,通过构建卷积神经网络(CNN)模型实现候选区域有效分类,最终精确确定飞机目标位置.最后,基于获取的真实遥感图像进行目标检测算法实验验证,经统计该算法的飞机目标检测率为94.38%,虚警率为3.76%,实验结果充分验证了该文算法的有效性,该算法可为机场监管、军事侦察等应用提供重要的技术支持.     

基于多尺度圆周频率滤波与卷积神经网络的遥感图像飞机目标检测方法研究

针对遥感图像飞机目标检测因目标尺度不一存在漏警、虚警等问题,该文基于遥感图像中飞机目标形状特征和灰度变化特点提出了一种多尺度圆周频率滤波(MSCFF)与卷积神经网络(CNN)相结合的MSCFF+CNN飞机目标自动检测算法。该算法首先采用多尺度圆周频率滤波器滤除遥感图像复杂背景,实现不同尺度飞机目标候选区域提取;然后,通过构建卷积神经网络(CNN)模型实现候选区域有效分类,最终精确确定飞机目标位置。最后,基于获取的真实遥感图像进行目标检测算法实验验证,经统计该算法的飞机目标检测率为94.38%,虚警率为3.76%,实验结果充分验证了该文算法的有效性,该算法可为机场监管、军事侦察等应用提供重要的技术支持。     

基于卷积神经网络的印刷电路板色环电阻检测与定位方法

色环电阻是印刷电路板(PCB)中最常用的电子元器件之一,主要依靠色环的排列顺序和颜色等视觉信息进行区分,易发生装配错误。但是色环电阻装配质量的人工检测方法效率低、误检率高,而传统的基于图像处理技术的自动检测方法鲁棒性较差,难以解决不同拍摄角度、物距及光照条件下的PCB板色环电阻检测问题。针对这一问题,该文提出一种基于卷积神经网络(CNN)的PCB板色环电阻自动检测与定位方法,首先采用编码器-解码器结构的卷积神经网络模型及带有权重的交叉熵损失函数的网络训练方法,较好地解决了复杂光照及场景下PCB板色环电阻的图像分割问题;然后采用最小面积外接矩形方法定位单个色环电阻,并通过仿射变换对色环电阻位置进行垂直校正;最后通过高斯模板匹配方法实现了色环电阻的色环定位。采用1270幅PCB图像对该文方法进行了实验和验证,并与传统的基于形态学和基于模板匹配的色环电阻检测方法进行了对比,结果表明,该文方法在召回率、准确率及重叠度等性能指标上具有明显优势,处理速度快,能满足实际应用要求。     

基于边缘检测与双边滤波的彩色图像去噪

 针对彩色图像双边滤波去噪方法存在的不足,本文提出一种边缘检测与双边滤波相结合的彩色图像去噪方法.首先利用细胞神经网络(CNN)模型导出一种新的彩色图像分块自适应边缘检测算法,继承了CNN灰度边缘检测算法定位准确的优点,又弥补了CNN现有算法不能直接处理彩色图像的空白.接下来提出一种针对图像增强的边缘滤波算法,通过两级边缘检测满足去噪不同阶段对边缘检测的不同要求.在此基础上,用改进的双边滤波器对彩色图像进行去噪,通过非抗噪边缘图对噪声范围进行定位,以缩小双边滤波的范围,减少去噪过程带来的图像模糊,并且对双边滤波加权平均方式进行改进,减小噪声点本身的权重,降低高频噪声的影响.最后根据滤波后的去噪边缘图对彩色图像进行增强.实验结果表明,文中方法在有效去除噪声的同时保护和增强了图像中的边缘.     

注意力引导梯形金字塔融合网络识别新冠肺炎X射线影像

新型冠状病毒肺炎(corona virus disease 2019,COVID-19)严重影响人类社会和经济的发展,威胁人类的健康。如何更准确、快速地排查感染病毒的患者,使用卷积神经网络(convolutional neural network, CNN)的方法识别COVID-19胸部X射线影像,完成计算机自动辅助诊断。但是,由于识别精度不高,难以准确判断是否感染了COVID-19。为了提高网络模型对COVID-19胸部X射线影像的识别性能,首先提出注意力引导梯形金字塔融合网络(attention steered trapezoid pyramid fusion network, ASTPNet),该网络可以附加在不同的CNN上,有效地利用模型中深层与浅层网络的特点;其次提出注意力引导块(attention steered block, AS Block),通过通道和空间注意力,强调通道和空间中的有效语义信息,弱化无效的干扰信息,高效地聚合加权信息。最终实验结果表明：将ASTPNet附加在VGG16/19、ResNet34/50和ResNeXt上,识别精度有了显著提升;应用于自建的C...     

A GAN based method for multiple prohibited items synthesis of X-ray security image

Detecting prohibited item based on convolutional neural networks(CNNs) is of great significance to ensure public safety. However, the natural occurrence of such prohibited items is a small-probability event, collecting enough datasets to support CNN training is a big challenge. In this paper, we propose a new method for synthesizing X-ray security image with multiple prohibited items from semantic label images basing on Generative Adversarial Networks(GANs). Theoretically, we can use it to synthesize as many X-ray images as needed. A new generator architecture with Res 2 Net is presented, which is more effective in learning multi-scale features of different prohibited items images. This method is extended by establishing the semantic label library which contains 14 000 images. So we totally synthesize 14 000 Xray security images. The experimental results show the super performance(Fréchet Inception Distance(FID) score of 30.55). And we achieve 0.825 of mean average precision(m AP) with Single Shot Multi Box Detector(SSD) for object detection, demonstrating the effectiveness of our approach.     

基于改进Faster R CNN的光伏组件红外热斑检测算法

光伏故障检测对光伏电站智能运维具有重要意义。针对光伏组件红外图像中热斑目标小、难检测的问题,研究了基于改进Faster R CNN的光伏组件红外热斑故障检测模型。将Swin Transformer作为Faster R CNN模型中的特征提取模块,捕获图像的全局信息,建立特征之间的依赖关系,提高模型的建模能力;进一步利用BiFPN进行特征融合,改善了热斑故障由于目标小和特征不明显容易被模型忽略掉的问题;同时为了抑制光伏红外图像中背景和噪声的干扰,加入轻量级注意力模块CBAM,使模型更加关注重要通道和关键区域,提高对热斑故障检测精度。在自建光伏组件图像数据集上进行实验,热斑故障检测精度高达915,验证了本文模型对光伏组件热斑故障检测的有效性。     

基于深度学习的远程视频监控异常图像检测

针对复杂场景下远程视频监控图像异常检测困难、传统算法功能单一(仅针对某种特定场景或某种异常图像进行检测)等问题,提出一种基于深度学习的全自动远程视频异常图像检测方法。首先采用Xavier方法对自行设计的卷积神经网络(Convolutional Neural Network,CNN)的参数进行初始化,然后将标准化后的视频差分图送入CNN的输入层,通过特征提取及下采样,最后在CNN的输出层获得远程视频异常图像检测结果。实验结果表明,该方法可以对远程视频监控中突然出现遮挡、模糊和场景切换等多种异常同时进行实时在线检测,准确率可达88.75%。     

一种基于修正激活函数的CNN车载毫米波雷达目标检测方法

为了提高车载毫米波雷达在复杂城市道路环境中目标检测的抗杂波与干扰能力,本文利用卷积神经网络（CNN）特征参数提取和目标分类特性,提出了一种改进的基于CNN的车载毫米波雷达目标检测方法。该方法首先将毫米波雷达回波信号距离-多普勒二维数据运用滑窗进行分割,并采用CNN网络模型处理分割后的二维矩阵,训练二维CNN网络模型及其参数,使其具有提取回波特征并基于特征参数模型进行目标分类的能力,从而实现目标检测功能。通过对卷积神经网络模型结构进行优化,增加批量归一化层,优化Dropout层使得低权重特征失活,自适应地删减部分神经元节点修正该层非线性激活函数,进一步降低了CNN模型目标检测的虚警概率。实验结果表明,在相同虚警概率条件下,CNN网络检测方法目标发现概率优于传统的单元平均恒虚警检测方法,并且在低信噪比的条件下仍然能够保持较高的发现概率;在同等发现概率水平下,修正后CNN网络检测方法的虚警概率较修正前可提高约1个数量级。     

基于多层次特征差异图的视觉场景识别

场景外观剧烈变化引起的感知偏差和感知变异给视觉场景识别带来了很大的挑战。现有的利用卷积神经网络（CNN）的视觉场景识别方法大多数直接采用CNN特征的距离并设置阈值来衡量两幅图像之间的相似性,当场景外观剧烈变化时效果较差,为此提出了一种新的基于多层次特征差异图的视觉场景识别方法。首先,一个在场景侧重的数据集上预训练的CNN模型被用来对同一场景中感知变异的图像和不同场景中感知偏差的图像进行特征提取。然后,根据CNN不同层特征具有的不同特性,融合多层CNN特征构建多层次特征差异图来表征两幅图像之间的差异。最后,视觉场景识别被看作二分类问题,利用特征差异图训练一个新的CNN分类模型来判断两幅图像是否来自同一场景。实验结果表明,由多层CNN特征构建的特征差异图能很好地反映两幅图像之间的差异,文中提出的方法能有效地克服感知偏差和感知变异,在场景外观剧烈变化下取得很好的识别效果。     

基于细胞神经网络的肝脏核磁共振图像自动分割方案

Currently, the processing speed of existing automatic liver segmentation for Magnetic Resonance Imaging (MRI) images is relatively slow. An automatic liver segmentation scheme for MRI images based on Cellular Neural Networks (CNN) is presented in this paper. It ensures the validity of this scheme and at the same time completes the image segmentation faster to accurately calculate the liver volume by using parallel computing in real time. In order to facilitate the CNN image processing, firstly, three-dimensional liver MRI images should be transformed into binary images; secondly, an appropriate template parameter of the Global Connectivity Detection CNN (GCD CNN) shall be selected to probe the connectivity of the liver to extract the entire liver; and then the Hole-Filler CNN (HF CNN) are used to repair the entire extracting liver and improve the accuracy of liver segmentation; finally, the liver volume is obtained. Results show that the scheme can ensure the accuracy of the automatic segmentation of the liver, and it can also improve the processing speed at the same time. The liver volume calculated is in line with the clinical diagnosis.     

基于深度学习的激光剪切散斑干涉无损检测缺陷识别

剪切散斑干涉技术通过测量物体表面变形来推断其内部缺陷,具有高灵敏度、检测范围广、精度高等优点,是一种极具潜力的复合材料无损检测技术。目前缺陷识别主要采用人工方式,而人工识别不但检测效率低且受到专业性限制。为了提高剪切散斑干涉无损检测方法中的缺陷识别精度和效率,本文提出基于深度学习剪切散斑干涉缺陷识别方法。利用高精度四步相移技术获取剪切散斑相位条纹高质量成像;引入了应用广泛的YOLOv5和Faster R-CNN目标检测算法,通过实验采集了大量的缺陷图像,分别用YOLOv5和Faster R-CNN两种算法获得训练模型。然后将这两种模型分别应用于剪切散斑干涉无损检测中的复合材料缺陷检测。最后,实验从检测速率和检测精度方面对模型识别效果进行了对比分析。实验结果表明,激光剪切散斑干涉技术结合深度学习的方法能有效地实现剪切散斑干涉无损检测的缺陷自动识别,Faster R-CNN和YOLOv5的检测速率分别能达到11 f/s和50 f/s,并且两种深度学习算法的平均精度均能达到92%以上,验证了提出方法的可行性。     

基于CNN的泄漏电缆入侵检测定位算法

泄漏电缆入侵检测系统所处的外部环境较为复杂,为降低环境因素对泄露电缆入侵检测的影响,提出了基于卷积神经网络的入侵检测算法。通过卷积神经网络处理大量的样本数据,并从数据中自动提取内在特性,实现泄漏电缆电磁入侵检测系统更低的误报率、漏报率和更高的定位精度的目标,搭建了卷积神经网络入侵检测模型,并用样本数据对模型进行训练和测试。模型测试结果表示其具有低漏报率和误报率,定位精度可达到1 m。     

基于深度学习的海上目标一维序列信号目标检测方法

当前海面目标检测方法多基于统计理论,检测性能受背景统计特性假设的影响,本文从信号预测和特征分类两个角度,分别采用长短时记忆网络(LSTM)和卷积神经网络(CNN)对信号时间序列幅度信息进行处理,用于海上目标一维序列雷达信号检测,该方法不需事先假设背景统计特性,泛化能力更强。基于LSTM序列预测的目标检测方法通过用海杂波信号幅度时间序列对网络进行训练,再用训练后的网络对后续序列进行预测,并与后续实测信号进行比较,实现目标检测。基于CNN序列分类的目标检测方法中采用截取的海杂波信号和目标信号幅度序列作为数据集样本,对一维卷积核CNN进行训练,使其具有识别目标杂波信号特征能力,从而实现目标检测。最后,采用IPIX和CSIR实测海杂波数据对两种方法进行验证,结果表明两种方法均可实现一维序列信号中海面目标的检测,但LSTM预测方法对于长序列检测的实时性有待于进一步提高;CNN分类方法可实现实时检测,但仅利用信号幅度信息,检测性能仍需进一步提升。      

Low Latency Fog-Centric Deduplication Approach to Reduce IoT Healthcare Data Redundancy

In today’s highly computerized society, detection and recognition of text present in natural scene images is complex and difficult to be properly recognized by human vision. Most of the existing algorithms and models mainly focus on detection and recognition of text from still images. Many of the recent machine translation systems are built using the Encoder-Decoder framework which works on the format of encoding the sequence of input and then based on the encoded input, the output is decoded. Both the encoder and the decoder use an attention mechanism as an interface, making the model complex. Aiming at this situation, an alternative method for recognition of texts from videos is proposed. The proposed approach is based on a single Two-Dimensional Convolutional Neural Network (2D CNN). An algorithm for extracting features from an image called the crosswise feature extraction is also proposed. The proposed model is tested and shows that crosswise feature extraction gives better recognition accuracy by requiring a lesser period of time for training than the conventional feature extraction technique used by CNN.

     

Parallel design of convolutional neural networks for remotesensing images object recognition based ondata-driven array processor

Object recognition in very high-resolution remote sensing images is a basic problem in the field of aerial and satellite image analysis. With the development of sensor technology and aerospace remote sensing technology, thequality and quantity of remote sensing images are improved. Traditional recognition methods have a certainlimitation in describing higher-level features, but object recognition method based on convolutional neural network(CNN) can not only deal with large scale images, but also train features automatically with high efficiency. It ismainly used on object recognition for remote sensing images. In this paper, an AlexNet CNN model is trained using2 100 remote sensing images, and correction rate can reach 97.6% after 2 000 iterations. Then based on trainedmodel, a parallel design of CNN for remote sensing images object recognition based on data-driven array processor(DDAP) is proposed. The consuming cycles are counted. Simultaneously, the proposed architecture is realized onXilinx V6 development board, and synthesized based on SMIC 130 nm complementary metal oxid semiconductor(CMOS) technology. The experimental results show that the proposed architecture has a certain degree ofparallelism to achieve the purpose of accelerating calculations.     

Image descreening by GA-CNN-based texture classification

This work proposes an image-descreening technique based on texture classification using a cellular neural network (CNN) with template trained by genetic algorithm (GA), called GA-CNN. Instead of using the fixed filters for image descreening, we are equipped with a more pliable mechanism for classifications in screening patterns. Using CNN makes it possible to get an accurate texture classification result in a faster speed by its superiority of implementable hardware and the flexible choices of templates. The use of the GA here helps us to look for the most appropriate template for CNNs more adaptively and methodically. The evolved parameters in the template for CNNs can not only provide a quicker classification mechanism but also help us with a better texture classification for screening patterns. After the class of screening patterns in the querying images is determined by the trained GA-CNN-based texture classification system, the recommendatory filters are induced to solve the screening problems. The induction of the classification in screening patterns has simplified the choice of filters and made it valueless to determine a new structured filter. Eventually, our comprehensive methodology is going to be topped off with more desirable results and the indication for the decrease in time complexity.     

No-reference image quality assessment based on hybrid model

The aim of research on the no-reference image quality assessment problem is to design models that can predict the quality of distorted images consistently with human visual perception. Due to the little prior knowledge of the images, it is still a difficult problem. This paper proposes a computational algorithm based on hybrid model to automatically extract vision perception features from raw image patches. Convolutional neural network (CNN) and support vector regression (SVR) are combined for this purpose. In the hybrid model, the CNN is trained as an efficient feature extractor, and the SVR performs as the regression operator. Extensive experiments demonstrate very competitive quality prediction performance of the proposed method.     

基于改进VGG网络的单体热电池X光图像无损检测方法研究

为解决单体热电池生产中出现的安装错误、人工检测耗时耗力的问题,提出一个结合迁移学习和卷积 神 经网络(convolutional neural network,CNN) 的单体热电池缺陷检测模型。首先,对数据集图像进行裁剪、加噪等预处理,以VGG16(visual geometry group 16) 网络作为 模型的骨干架构,在瓶颈层后增添选择性核(selective kernel,SK) 卷积;然后,增添全局平均池化(global average pooling,GAP) 层, 增加Dropout层及添加 L2 正则化等微调操作,得到单体热电池缺陷检测模型Q-VGGNet;最后,在大型公开数据集ImageNet上进 行预训练学习,将获得的权重参数迁移到单体热电池图像识别模型Q-VGGNet上。测试实验表明:6种 网络模型对数据集缺陷图像的总体识别准确率分别达到了98.39%、94.44%、97.27%、96.34%、93.71%、 95.61%,Q-VGGNet网 络模型 对合格图像和 漏装负极、极耳断裂、漏装集流片3种缺陷图像 识别准确率 分别达到了99.6%,95.9%,99.6%和98.4%。检测结果表明:该方法能够更准确、快速地检测热电池缺陷, 拥有良好的缺陷诊断能力,较传统方法提高近3%,为人工检测单体热电池缺陷提供了良好的解决途径。