卷积神经网络和深度置信网络在SAR影像冰水分类的性能评估

目的 海冰分类是海冰监测的主要任务之一。目前基于合成孔径雷达SAR影像的海冰分类方法分为两类：一类是基于海冰物理特性与SAR成像特征等进行分类,这需要一定的专业背景;另一类基于传统的图像特征分类,需要人为设计特征,受限于先验知识。近年来深度学习在图像分类和目标识别方面取得了巨大的成功,为了提高海冰分类精度及海冰分类速度,本文尝试将卷积神经网络（CNN）和深度置信网络（DBN）用于海冰的冰水分类,评估不同类型深度学习模型在SAR影像海冰分类方面的性能及其影响因素。方法 首先根据加拿大海冰服务局（CIS）的冰蛋图构建海冰的冰水数据集;然后设计卷积神经网络和深度置信网络的网络架构;最后评估两种模型在不同训练样本尺寸、不同数据集大小和网络层数、不同冰水比例的测试影像以及不同中值滤波窗口的分类性能。结果 两种模型的总体分类准确率达到93%以上,Kappa系数0.8以上,根据分类结果得到的海冰区域密集度与CIS的冰蛋图海冰密集度数据一致。海冰的训练样本尺寸对分类结果影响显著,而训练集大小以及网络层数的影响较小。在本文的实验条件下,CNN和DBN网络的最佳分类样本尺寸分别是16×16像素和32×32像素。结论 利用CNN和DBN模型对SAR影像海冰冰水分类,并进行性能分析。发现深度学习模型用于SAR影像海冰分类具有潜力,与现有的海冰解译图的制作流程和信息量相比,基于深度学习模型的SAR影像海冰分类可以提供更加详细的海冰地理分布信息,并且减小时间和资源成本。     

面向移动端的轻量化卷积神经网络结构

随着移动设备的大量普及,将卷积神经网络应用于移动设备具有极大的实用价值。虽然随着技术的发展,目前移动设备的计算能力和存储资源都有了极大的提高,但是在移动设备上运行卷积神经网络仍然具有很大的挑战。为了解决这个问题,提出了一种轻量化的卷积神经网络结构S-MobileNet。该结构可以显著地减少网络模型的参数量以及降低模型的计算复杂度。为了全面评测S-MobileNet的性能,在CIFAR-10、CIFAR-100和ImageNet等图像分类数据集上进行了相关实验。实验结果表明,所提出的网络结构在保持MobileNetV2同等准确度的前提下,网络模型的参数量较MobileNetV2减少了近1/3,计算复杂度较MobileNetV2降低了近40%。     

The skin cancer classification using deep convolutional neural network

Multimedia Tools and Applications - This paper addresses the demand for an intelligent and rapid classification system of skin cancer using contemporary highly-efficient deep convolutional neural...     

Quantum convolutional neural network for image classification

Pattern Analysis and Applications - In this paper we propose two scale-inspired local feature extraction methods based on Quantum Convolutional Neural Network (QCNN) in the Tensorflow quantum...     

面向运动想象脑电图识别的镜卷积神经网络

目的 脑电图（electroencephalogram,EEG）是一种灵活、无创、非侵入式的大脑监测方法,广泛应用于运动想象脑机接口系统中,运动想象脑电图识别精度是决定系统性能的关键因素。然而由于脑电图采集时间长、个体差异大等原因,导致单个受试者可用于模型训练的样本数量少,严重影响了卷积神经网络在脑电图识别任务中的表现。为此,本文提出一种镜卷积神经网络（mirror convolutional neural network,MCNN）模型,使用集成学习与数据扩增方法提高运动想象脑电图识别精度。方法 在训练阶段,基于源脑电通过互换左右侧脑电通道构造镜像脑电,并与源脑电一起用于源卷积网络训练,有效扩增了训练样本;在预测阶段,复制已训练源卷积网络作为镜像卷积网络,将测试集中的源脑电输入源卷积网络,构造的镜像脑电输入镜像卷积网络,集成源卷积网络与镜像卷积网络输出的类别预测概率,形成最终类别预测。结果 为了验证模型的有效性和通用性,基于3种不同运动想象脑电图识别卷积网络模型分别构造镜卷积网络,并在第4届脑机接口大赛2a与2b数据集上进行实验验证。实验结果与原始模型相比,运动想象四分类和二分类准确率分别平均提高了4.83%和4.61%,显著提高了识别精度。结论 本文面向运动想象脑电图识别,提出了镜卷积神经网络模型,通过集成学习与数据扩增方法提高运动想象识别精度,有效改善了运动想象脑机接口性能。     

Bi-branch deconvolution-based convolutional neural network for image classification

With the rise of deep neural network, convolutional neural networks show superior performances on many different computer vision recognition tasks. The convolution is used as one of the most efficient ways for extracting the details features of an image, while the deconvolution is mostly used for semantic segmentation and significance detection to obtain the contour information of the image and rarely used for image classification. In this paper, we propose a novel network named bi-branch deconvolution-based convolutional neural network (BB-deconvNet), which is constructed by mainly stacking a proposed simple module named Zoom. The Zoom module has two branches to extract multi-scale features from the same feature map. Especially, the deconvolution is borrowed to one of the branches, which can provide distinct features differently from regular convolution through the zoom of learned feature maps. To verify the effectiveness of the proposed network, we conduct several experiments on three object classification benchmarks (CIFAR-10, CIFAR-100, SVHN). The BB-deconvNet shows encouraging performances compared with other state-of-the-art deep CNNs.     

利用CNN和PCA约束优化模型实现稀疏表示分类

目的 传统的稀疏表示分类方法运用高维数据提升算法的稀疏分类能力,早已引起了广泛关注,但其忽视了测试样本与训练样本间的信息冗余,导致了不确定性的决策分类问题。为此,本文提出一种基于卷积神经网络和PCA约束优化模型的稀疏表示分类方法（EPCNN-SRC）。方法 首先通过深度卷积神经网络计算,在输出层提取对应的特征图像,用以表征原始样本的鲁棒人脸特征。然后在此特征基础上,构建一个PCA（principal component analysis）约束优化模型来线性表示测试样本,计算对应的PCA系数。最后使用稀疏表示分类算法重构测试样本与每类训练样本的PCA系数来完成分类。结果 本文设计的分类模型与一些典型的稀疏分类方法相比,取得了更好的分类性能,在AR、FERET、FRGC和LFW人脸数据库上的实验结果显示,当每类仅有一个训练样本时,EPCNN-SRC算法的识别率分别达到96.92%、96.15%、86.94%和42.44%,均高于传统的表示分类方法,充分验证了本文算法的有效性。同时,本文方法不仅提升了对测试样本稀疏表示的鲁棒性,而且在保证识别率的基础上,有效降低了算法的时间复杂度,在FERET数据库上的运行时间为4.92 s,均低于一些传统方法的运行时间。结论 基于卷积神经网络和PCA约束优化模型的稀疏表示分类方法,将深度学习特征与PCA方法相结合,不仅具有较好的识别准确度,而且对稀疏分类也具有很好的鲁棒性,尤其在小样本问题上优势显著。     

Brain tumor classification using deep convolutional autoencoder-based neural network: multi-task approach

Diagnosis, detection and classification of tumors, in the brain MRI images, are important because misdiagnosis can lead to death. This paper proposes a method that can diagnose brain tumors in the MRI images and classify them into 5 categories using a Convolutional Neural Network (CNN). The proposed network uses a Convolutional Auto-Encoder Neural Network (CANN) to extract and learn deep features of input images. Extracted deep features from each level are combined to make desirable features and improve results. To classify brain tumor into three categories (Meningioma, Glioma, and Pituitary) the proposed method was applied on Cheng dataset and has reached a considerable performance accuracy of 99.3%. To diagnosis and grading Glioma tumors, the proposed method was applied on IXI and BraTS 2017 datasets, and to classify brain images into six classes including Meningioma, Pituitary, Astrocytoma, High-Grade Glioma, Low-Grade Glioma and Normal images (No tumor), the all datasets including IXI, BraTS2017, Cheng and Hazrat-e-Rassol, was used by the proposed network, and it has reached desirable performance accuracy of 99.1% and 98.5%, respectively.

     

用于肺炎图像分类的优化卷积神经网络方法

目前,卷积神经网络（CNN）开始应用在肺炎分类领域。针对层数较浅、结构较为简单的卷积网络对肺炎识别的准确率难以提高的情况,采用深度学习方法,并针对采用深度学习方法时常常需要消耗大量的系统资源,导致卷积网络难以在用户端部署的问题,提出一种使用优化的卷积神经网络的分类方法。首先,根据肺炎图像的特征,选择具有良好图像分类性能的AlexNet与InceptionV3模型;然后,利用医学影像特点对层次更深、结构更加复杂的InceptionV3模型进行预训练;最后,通过知识蒸馏的方法,将训练好的"知识"（有效信息）提取到AlexNet模型中,从而实现在减少系统资源占用的同时,提高准确率的效果。实验数据表明,使用知识蒸馏后,AlexNet模型的准确率、特异性与灵敏度分别提高了4.1、7.45、1.97个百分点,且对图像处理器（GPU）占用相比InceptionV3模型减小了51个百分点。     

CT image classification based on convolutional neural network

Neural Computing and Applications - With the rapid development of the Internet, image information is explosively growing. Traditional image classification methods are difficult to deal with huge...     

采用卷积神经网络的低风险可行地貌分类方法

针对现有识别方法中风险地貌误判率高、手动地貌特征提取具有局限性等问题, 提出了用于室外移动机器 人的低风险地貌识别策略. 该策略以降低移动机器人遇险率为高优先级目标, 采用双重验证策略, 首先采用多分类 器对所有地貌进行识别, 其后使用二分类器对多分类结果中的安全地貌再次鉴别. 基于该策略, 分别设计了2个卷积 神经网络(CNN), Terrain–CNNⅠ用于多分类识别, Terrain–CNNⅡ则用于二分类安全确认. 为解决地貌样本相对稀缺 问题, 收集了包含水面、草地、泥地、柏油路、沙地、碎石路共6类地貌图像, 通过数据增强方式快速扩充数据集用于 网络的训练与测试. 实验结果表明: 所述方法在维持整体地貌识别率很高的前提下, 显著降低了关键危险地貌的误 判率.     

基于双流对称特征融合网络模型的海洋船舶目标识别

海洋船舶目标识别在民用和军事领域有着重要的战略意义, 本文针对可见光图像和红外图像提出了一种 基于注意力机制的双流对称特征融合网络模型, 以提升复杂感知环境下船舶目标综合识别性能. 该模型利用双流对 称网络并行提取可见光和红外图像特征, 通过构建基于级联平均融合的多级融合层, 有效地利用可见光和红外两种 模态的互补信息获取更加全面的船舶特征描述. 同时将空间注意力机制引入特征融合模块, 增强融合特征图中关 键区域的响应, 进一步提升模型整体识别性能. 在VAIS实际数据集上进行系列实验证明了该模型的有效性, 其识别 精确度能达到87.24%, 综合性能显著优于现有方法.     

Active weighted mapping-based residual convolutional neural network for image classification

Multimedia Tools and Applications - In visual recognition, the key to the performance improvement of ResNet is the success in establishing the stack of deep sequential convolutional layers using...     

基于特征整合的卷积神经网络草地分类算法

为提高遥感影像草地分类的精度,分析了卷积神经网络中提取图像特征的特点,提出了一种基于特征整合深度神经网络的遥感影像特征提取算法。首先,将遥感影像数据进行PCA白化处理,降低数据之间的相关性,加快神经网络学习的速率;其次,将从卷积神经网络中提取到的浅层特征和深层特征进行双线性整合,使得整合后的新特征更加完善和优化;最后,对遥感数据进行训练,由于新特征中有效信息的增加,使得特征表达能力得到提高,达到提高草地分类准确率的目的。实验结果表明:该算法能够有效地提高草地分类的准确率,分类精度达到94.65%,相较于卷积神经网络、BP神经网络和基于SVM的分类算法分别提高了4.3%、10.39%和15.33%。     

Enhanced brain tumor classification using an optimized multi-layered convolutional neural network architecture

Multimedia Tools and Applications - Detecting and classifying a brain tumor is a challenge that consumes a radiologist’s time and effort while requiring professional expertise. To resolve...     

Mental imagery for a conversational robot

To build robots that engage in fluid face-to-face spoken conversations with people, robots must have ways to connect what they say to what they see. A critical aspect of how language connects to vision is that language encodes points of view. The meaning of my left and your left differs due to an implied shift of visual perspective. The connection of language to vision also relies on object permanence. We can talk about things that are not in view. For a robot to participate in situated spoken dialog, it must have the capacity to imagine shifts of perspective, and it must maintain object permanence. We present a set of representations and procedures that enable a robotic manipulator to maintain a "mental model" of its physical environment by coupling active vision to physical simulation. Within this model, "imagined" views can be generated from arbitrary perspectives, providing the basis for situated language comprehension and production. An initial application of mental imagery for spatial language understanding for an interactive robot is described.     

Improved convolutional neural network in remote sensing image classification

Neural Computing and Applications - The classification of land cover is the first step in the analysis and application of remote sensing data in land resources. How to solve the multi-category...     

A central-point-enhanced convolutional neural network for high-resolution remote-sensing image classification

As one of the most important algorithms in the field of deep learning technology, the convolutional neural network (CNN) has been successfully applied in many fields. CNNs can recognize objects in an image by considering morphology and structure rather than simply individual pixels. One advantage of CNNs is that they exhibit translational invariance; when an image contains a certain degree of distortion or shift, a CNN can still recognize the object in the image. However, this advantage becomes a disadvantage when CNNs are applied to pixel-based classification of remote-sensing images, because their translational invariance characteristics causes distortions in land-cover boundaries and outlines in the classification result image. This problem severely limits the application of CNNs in remote-sensing classification. To solve this problem, we propose a central-point-enhanced convolutional neural network (CE-CNN) to classify high-resolution remote-sensing images. By introducing the central-point-enhanced layer when classifying a sample, the CE-CNN increases the weight of the central point in feather maps while preserving the original textures and characteristics. In our experiment, we selected four representative positions on a high-resolution remote-sensing image to test the classification ability of the proposed method and compared the CE-CNN with the traditional multi-layer perceptron (MLP) and a traditional CNN. The results show that the proposed method can not only achieves a higher classification accuracy but also less distortion and fewer incorrect results at the boundaries of land covers. We further compared the CE-CNN with six state-of-the-art methods: k-NN, maximum likelihood, classification and regression tree (CART), MLP, support vector machine, and CNN. The results show that the CE-CNN’s classification accuracy is better than the other methods.     

基于改进的一维卷积神经网络的异常流量检测

针对传统机器学习方法对特征依赖大、基于深度学习的检测方法效率低以及易过拟合的问题,提出一种基于改进的一维卷积神经网络（ICNN-1D）的异常流量检测方法（AFM-ICNN-1D）。与传统卷积神经网络（CNN）采用的“卷积-池化-全连接”结构不同,ICNN-1D主要由2个卷积层、2个全局池化层、1个dropout层和1个全连接输出层构成;其次,将预处理后的数据输入到ICNN-1D中,并将经过两次卷积之后的结果作为全局平均池化层与全局最大池化层的输入,之后将所得到的输出数据进行合并再送入全连接层进行分类;最后根据分类结果与真实数据对网络模型进行调优,再将训练好的模型用于异常流量检测。CIC-IDS-2017数据集上的实验结果显示,AFM-ICNN-1D的精准率和召回率均达到了98%,优于对比的k近邻（kNN）和随机森林（RF）方法;而且与传统的CNN相比,该方法的参数减少了约97%,训练时间缩短了约40%。实验结果表明,AFM-ICNN-1D具有较高的检测性能,能减少训练时间、避免过拟合现象的发生,而且能更好地保留流量数据的局部特征。     

Training strategies for neural network soft classification of remotely-sensed imagery

Recently the 'soft' classification approach has gained in popularity against the discrete (hard) classification of land cover from remotely-sensed imagery. An empirical study is presented to test training procedures with neural networks for soft (mixture) classification. The results show thatland cover mixtures are best recognized following training with two-component mixed pixels, and that linearly re-scaled or binned target vector representations are equally satisfactory. Interestingly, dominant classes within pixels are also better recognized by training with wider varieties of class mixtures.