An object-based and heterogeneous segment filter convolutional neural network for high-resolution remote sensing image classification

In recent years, object-based segmentation methods and shallow-model classification algorithms have been widely integrated for remote sensing image supervised classification. However, as the image resolution increases, remote sensing images contain increasingly complex characteristics, leading to higher intraclass heterogeneity and interclass homogeneity and thus posing substantial challenges for the application of segmentation methods and shallow-model classification algorithms. As important methods of deep learning technology, convolutional neural networks (CNNs) can hierarchically extract higher-level spatial features from images, providing CNNs with a more powerful recognition ability for target detection and scene classification in high-resolution remote sensing images. However, the input of the traditional CNN is an image patch, the shape of which is scarcely consistent with a given segment. This inconsistency may lead to errors when directly using CNNs in object-based remote sensing classification: jagged errors may appear along the land cover boundaries, and some land cover areas may overexpand or shrink, leading to many obvious classification errors in the resulting image. To address the above problem, this paper proposes an object-based and heterogeneous segment filter convolutional neural network (OHSF-CNN) for high-resolution remote sensing image classi?cation. Before the CNN processes an image patch, the OHSF-CNN includes a heterogeneous segment filter (HSF) to process the input image. For the segments in the image patch that are obviously different from the segment to be classified, the HSF can differentiate them and reduce their negative influence on the CNN training and decision-making processes. Experimental results show that the OHSF-CNN not only can take full advantage of the recognition capabilities of deep learning methods but also can effectively avoid the jagged errors along land cover boundaries and the expansion/shrinkage of land cover areas originating from traditional CNN structures. Moreover, compared with the traditional methods, the proposed OHSF-CNN can achieve higher classification accuracy. Furthermore, the OHSF-CNN algorithm can serve as a bridge between deep learning technology and object-based segmentation algorithms thereby enabling the application of object-based segmentation methods to more complex high-resolution remote sensing images.     

Convolutional neural network based heterogeneous transfer learning for remote-sensing scene classification

ABSTRACT

Deep convolutional neural network (CNN) transfer has recently shown strong performance in scene classification of high-resolution remote-sensing images. However, the majority of transfer learning solutions are categorized as homogeneous transfer learning, which ignores differences between target and source domains. In this paper, we propose a heterogeneous model to transfer CNNs to remote-sensing scene classification to correct input feature differences between target and source datasets. First, we extract filters from source images using the principal component analysis (PCA) method. Next, we convolute the target images with the extracted PCA filters to obtain an adopted target dataset. Then, a pretrained CNN is transferred to the adopted target dataset as a feature extractor. Finally, a classifier is used to accomplish remote-sensing scene classification. We conducted extensive experiments on the UC Merced dataset, the Brazilian coffee scene dataset and the Aerial Images Dataset to verify the effectiveness of the proposed heterogeneous model. The experimental results show that the proposed heterogeneous model outperforms the homogeneous model that uses pretrained CNNs as feature extractors by a wide margin and gains similar accuracies by fine-tuning a homogeneous transfer learning model with few training iterations.     

1D convolutional neural networks for chart pattern classification in financial time series

This paper proposes a novel deep learning-based approach for financial chart patterns classification. Convolutional neural networks (CNNs) have made notable achievements in image recognition and computer vision applications. These networks are usually based on two-dimensional convolutional neural networks (2D CNNs). In this paper, we describe the design and implementation of one-dimensional convolutional neural networks (1D CNNs) for the classification of chart patterns from financial time series. The proposed 1D CNN model is compared against support vector machine, extreme learning machine, long short-term memory, rule-based and dynamic time warping. Experimental results on synthetic datasets reveal that the accuracy of 1D CNN is highest among all the methods evaluated. Results on real datasets also reveal that chart patterns identified by 1D CNN are also the most recognized instances when they are compared to those classified by other methods.

     

基于卷积神经网络的遥感图像分类研究

遥感图像分类是模式识别技术在遥感领域的具体应用,针对遥感图像处理中的分类问题,提出了一种基于卷积神经网络(convolutional neural networks,CNN)的遥感图像分类方法,并针对单源特征无法提供有效信息的问题,设计了一种多源多特征融合的方法,将遥感图像的光谱特征、纹理特征、空间结构特征等按空间维度以向量或矩阵的形式进行有效融合,以此训练CNN模型。实验表明,多源多特征相融合能够加快模型收敛速度,有效提高遥感图像的分类精度;与其他分类方法相比,CNN能够取得更高的分类精度,获得更优的分类效果。     

Land-use scene classification based on a CNN using a constrained extreme learning machine

ABSTRACT

Classifying land-use scenes from high-resolution remote-sensing imagery with high quality and accuracy is of paramount interest for science and land management applications. In this article, we proposed a new model for land-use scene classification by integrating the recent success of convolutional neural network (CNN) and constrained extreme learning machine (CELM). In the model, the fully connected layers of a pretrained CNN have been removed. Then, CNN works as a deep and robust convolutional feature extractor. After normalization, deep convolutional features are fed to the CELM classifier. To analyse the performance, the proposed method has been evaluated on two challenging high-resolution data sets: (1) the aerial image data set consisting of 30 different aerial scene categories with sub-metre resolution and (2) a Sydney data set that is a large high spatial resolution satellite image. Experimental results show that the CNN-CELM model improves the generalization ability and reduces the training time compared to state-of-the-art methods.     

Zernike-CNNs for image preprocessing and classification in printed register detection

In the register detection of printing field, a new approach based on Zernike-CNNs is proposed. The edge feature of image is extracted by Zernike moments (ZMs), and a recursive algorithm of ZMs called Kintner method is derived. An improved convolutional neural networks (CNNs) are investigated to improve the accuracy of classification. Based on the classic convolutional neural network (CNN), the improved CNNs adopt parallel CNN to enhance local features, and adopt auxiliary classification part to modify classification layer weights. A printed image is trained with 7?×?400 samples and tested with 7?×?100 samples, and then the method in this paper is compared with other methods. In image processing, Zernike is compared with Sobel method, Laplacian of Gaussian (LoG) method, Smallest Univalue Segment Assimilating Nucleus (SUSAN) method, Finite Impusle Response (FIR) method, Multi-scale Morphological Gradient (MMG) method. In image classification, improved CNNs are compared with classical CNN. The experimental results show that Zernike-CNNs have the best performance, the mean square error (MSE) of the training samples reaches 0.0143, and the detection accuracy of training samples and test samples reached 91.43% and 94.85% respectively. The experiments reveal that Zernike-CNNs are a feasible approach for register detection.

     

Tripartite-structure transformer for hyperspectral image classification

Hyperspectral images contain rich spatial and spectral information, which provides a strong basis for distinguishing different land-cover objects. Therefore, hyperspectral image (HSI) classification has been a hot research topic. With the advent of deep learning, convolutional neural networks (CNNs) have become a popular method for hyperspectral image classification. However, convolutional neural network (CNN) has strong local feature extraction ability but cannot deal with long-distance dependence well. Vision Transformer (ViT) is a recent development that can address this limitation, but it is not effective in extracting local features and has low computational efficiency. To overcome these drawbacks, we propose a hybrid classification network that combines the strengths of both CNN and ViT, names Spatial-Spectral Former(SSF). The shallow layer employs 3D convolution to extract local features and reduce data dimensions. The deep layer employs a spectral-spatial transformer module for global feature extraction and information enhancement in spectral and spatial dimensions. Our proposed model achieves promising results on widely used public HSI datasets compared to other deep learning methods, including CNN, ViT, and hybrid models.     

A subclass supported convolutional neural network for object detection and localization in remote-sensing images

This article proposes a novel subclass-based classifier based on convolutional neural networks (CNNs) for detecting objects more accurately on remote-sensing images. The proposed classifier, called subclass supported CNN (SSCNN), is used to separate the representation of the objects into subclasses such as nearcentre, centre, and border depending on the distance of the object centre to obtain more effective feature extractor. A three-stage object recognition framework is used to evaluate the performance of the proposed classifier. In the first of these stages, the Selective Search algorithm generates object proposals from the image. Then, the proposed SSCNN classifies the proposals. Finally, subclass-based localization evaluation function has been proposed to calculate the localization of the object with classification results. Due to the limited number of satellite image samples, pretrained AlexNet is used by transfer learning approach to build effective feature extractor. The proposed method has been compared with region-based CNN (R-CNN) on a four-class remote-sensing test dataset consisting of 411 airplanes, 240 baseball diamonds, 468 storage tanks, and 83 ground track fields. In addition, Faster R-CNN has been trained with SSCNN features and the performances of the trained Faster R-CNNs are comparatively evaluated on 10-class remote-sensing image dataset. Experiment results have shown that the proposed framework can locate the objects precisely.     

多媒体工程：2016——图像检索研究进展与发展趋势

目的 基于内容的图像检索方法利用从图像提取的特征进行检索,以较小的时空开销尽可能准确的找到与查询图片相似的图片。方法 本文从浅层特征、深层特征和特征融合3个方面对图像检索国内外研究进展和面临的挑战进行介绍,并对未来的发展趋势进行展望。结果 尺度下不变特征转换（SIFT）存在缺乏空间几何信息和颜色信息,高层语义的表达不够等问题;而CNN （convolutional neural network）特征则往往缺乏足够的底层信息。为了丰富描述符的信息,通常将SIFT与CNN等特征进行融合。融合方式主要包括：串连、核融合、图融合、索引层次融合和得分层（score-level）融合。"融合"可以有效地利用不同特征的互补性,提高检索的准确率。结论 与SIFT相比,CNN特征的通用性及几何不变性都不够强,依然是图像检索领域面临的挑战。     

Detection and classification of lung nodules in chest X-ray images using deep convolutional neural networks

Lung nodule classification is one of the main topics related to computer-aided detection systems. Although convolutional neural networks (CNNs) have been demonstrated to perform well on many tasks, there are few explorations of their use for classifying lung nodules in chest X-ray (CXR) images. In this work, we proposed and analyzed a pipeline for detecting lung nodules in CXR images that includes lung area segmentation, potential nodule localization, and nodule candidate classification. We presented a method for classifying nodule candidates with a CNN trained from the scratch. The effectiveness of our method relies on the selection of data augmentation parameters, the design of a specialized CNN architecture, the use of dropout regularization on the network, inclusive in convolutional layers, and addressing the lack of nodule samples compared to background samples balancing mini-batches on each stochastic gradient descent iteration. All model selection decisions were taken using a CXR subset of the Lung Image Database Consortium and Image Database Resource Initiative dataset separately. Thus, we used all images with nodules in the Japanese Society of Radiological Technology dataset for evaluation. Our experiments showed that CNNs were capable of achieving competitive results when compared to state-of-the-art methods. Our proposal obtained an area under the free-response receiver operating characteristic curve of 7.76 considering 10 false positives per image (FPPI), and sensitivity values of 73.1% and 79.6% with 2 and 5 FPPI, respectively.     

Fine-tuning deep convolutional neural networks for distinguishing illustrations from photographs

Systems for aggregating illustrations require a function for automatically distinguishing illustrations from photographs as they crawl the network to collect images. A previous attempt to implement this functionality by designing basic features that were deemed useful for classification achieved an accuracy of only about 58%. On the other hand, deep neural networks had been successful in computer vision tasks, and convolutional neural networks (CNNs) had performed good at extracting such useful image features automatically. We evaluated alternative methods to implement this classification functionality with focus on deep neural networks. As the result of experiments, the method that fine-tuned deep convolutional neural network (DCNN) acquired 96.8% accuracy, outperforming the other models including the custom CNN models that were trained from scratch. We conclude that DCNN with fine-tuning is the best method for implementing a function for automatically distinguishing illustrations from photographs.     

遥感图像飞机目标分类的卷积神经网络方法

目的 遥感图像飞机目标分类,利用可见光遥感图像对飞机类型进行有效区分,对提供军事作战信息有重要意义。针对该问题,目前存在一些传统机器学习方法,但这些方法需人工提取特征,且难以适应真实遥感图像的复杂背景。近年来,深度卷积神经网络方法兴起,网络能自动学习图像特征且泛化能力强,在计算机视觉各领域应用广泛。但深度卷积神经网络在遥感图像飞机分类问题上应用少见。本文旨在将深度卷积神经网络应用于遥感图像飞机目标分类问题。方法 在缺乏公开数据集的情况下,收集了真实可见光遥感图像中的8种飞机数据,按大致4∶1的比例分为训练集和测试集,并对训练集进行合理扩充。然后针对遥感图像与飞机分类的特殊性,结合深度学习卷积神经网络相关理论,有的放矢地设计了一个5层卷积神经网络。结果 首先,在逐步扩充的训练集上分别训练该卷积神经网络,并分别用同一测试集进行测试,实验表明训练集扩充有利于网络训练,测试准确率从72.4%提升至97.2%。在扩充后训练集上,分别对经典传统机器学习方法、经典卷积神经网络LeNet-5和本文设计的卷积神经网络进行训练,并在同一测试集上测试,实验表明该卷积神经网络的分类准确率高于其他两种方法,最终能在测试集上达到97.2%的准确率,其余两者准确率分别为82.3%、88.7%。结论 在少见使用深度卷积神经网络的遥感图像飞机目标分类问题上,本文设计了一个5层卷积神经网络加以应用。实验结果表明,该网络能适应图像场景,自动学习特征,分类效果良好。     

Addressing image and Poisson noise deconvolution problem using deep learning approaches

Digital images are more important in numerous contemporary applications, and the need for images in the technical field is also increasing drastically. It is used to recognize signatures and faces in many industries and is applicable for intelligent departments. The images are usually associated with the noise content; this may happen due to the instrument imperfections, troubleshooting while collecting data from the acquisition process, and another natural phenomenon. Poisson noise, also known as photon noise, is caused in the images due to the statistical essence of electromagnetic waves. X-ray, visible light, and gamma rays are electromagnetic waves. The enhancement of the convolution model in addressing images is challenging due to the various constituents such as optical aberrations, noise level, and optical setup. The modeling configuration of the image is attained using the point spread function (PSF), which is responsible for the system's impulse response. The quality image is retrieved by denoising and super-resolution (SR) methods; these methods simultaneously eliminate the noise content from the images. A Richardson–Lucy and alternating direction method of multipliers type of non-blind iterative algorithmic approaches associated with the PSF performance in addressing image is comparatively analyzed. The deep learning approach, convolutional neural networks (CNNs), is also employed to understand the nonlinear mapping relationship between the observed data and ground reality. The performance of the various network approaches is compared in this article. The result obtained shows that the deep learning CNNs achieved higher accuracy in producing denoising images. The goal of the proposed system model is to remove the interference noise in images. The high-resolution images are obtained by implementing a SR-based CNN model.     

Learning discriminative spatiotemporal features for precise crop classification from multi-temporal satellite images

ABSTRACT

Precise crop classification from multi-temporal remote sensing images has important applications such as yield estimation and food transportation planning. However, the mainstream convolutional neural networks based on 2D convolution collapse the time series information. In this study, a 3D fully convolutional neural network (FCN) embedded with a global pooling module and channel attention modules is proposed to extract discriminative spatiotemporal presentations of different types of crops from multi-temporal high-resolution satellite images. Firstly, a novel 3D FCN structure is introduced to replace 2D FCNs as well as to improve current 3D convolutional neural networks (CNNs) by providing a mean to learn distinctive spatiotemporal representations of each crop type from the reshaped multi-temporal images. Secondly, to strengthen the learning significance of the spatiotemporal representations, our approach includes 3D channel attention modules, which regulate the between-channel consistency of the features from the encoder and the decoder, and a 3D global pooling module, which selects the most distinctive features at the top of the encoder. Experiments were conducted using two data sets with different types of crops and time spans. Our results show that our method outperformed in both accuracy and efficiency, several mainstream 2D FCNs as well as a recent 3D CNN designed for crop classification. The experimental data and source code are made openly available at http://study.rsgis.whu.edu.cn/pages/download/.     

Local image quality measurement for multi-scale forensic palmprints

Numerous studies show that palmprint image quality has a significant effect on every stage of a palmprint recognition system. Although some palmprint image quality measurement(PIQM) methods are proposed, some insufficiency in classification accuracy occurs and attention to detail in measuring local area image quality of multi-scale palmprint images is lacking. On the one hand, the classification accuracy is not very high for 2-class classification and it degrades significantly as the number of classes increases. On the other hand, local area image quality measurement of multi-scale palmprint images has not yet been resolved since the handcrafted features designed through domain knowledge usually works for certain scale image blocks. Meanwhile, the intricate domain knowledge used in the previous methods is difficult for some common users to acquire. In this paper, we propose an end-to-end deep-learning method of strengthening representation ability that learns more abstract, essential, and reliable features to measure the local image quality for multi-scale forensic palmprints. Popular convolutional neural networks (CNNs) are considered because of their powerful representation ability in learning complex features. However, the powerful existing CNNs usually have complex architectures with a large amount of parameters, which need the support of high-performance computers. They are not suitable to be used directly for palmprint image quality assignment and the follow-up palmprint recognition work, which prefers real-time response on commonly available personal computers or even mobile devices. Hence, a new lightweight CNN must be designed to achieve a trade-off between high classification accuracy and practical usability. Considering the attributes of under-processed input images, we reduce the weight of the CNN architecture by reducing the amount of some parameters, and finally a lightweight CNN is designed. As a result, a raw rectangular palmprint image of variable size can be put into the trained model directly and a quality label quickly predicted with high accuracy. After comparison with previous methods, results show that the proposed method can deal with un-pre-processed raw images of a multi-scale input size. Furthermore, it can acquire a richer amount of quality classes with a higher accuracy, which are stable on many different datasets. It also leads to finer and more precise full palmprint image quality maps when compared to previous methods.

     

基于CNN的高分遥感影像深度语义特征提取研究综述

近年来,深度学习作为计算机视觉的研究热点,在诸多方面得以发展与应用。特征提取是理解和分析高分遥感影像的关键基础。为促进高分遥感影像特征提取技术的发展,总结了深度学习模型在高分遥感影像特征提取技术的研究与发展,如:AlexNet,VGG-网和GoogleNet等卷积网络模型在深度语义特征提取中的应用。此外,重点分析和讨论了以卷积神经网络模型为基础的各类深度学习模型在高分遥感影像特征提取方面的应用与创新,如:迁移学习的应用;卷积神经网络(Convolutional Neural Network,CNN)模型结构的改变;CNN模型与其他模型结构的结合等方式,均提升了深度语义特征提取能力。最后,对卷积神经网络模型在高分遥感影像深度语义特征提取方面存在的问题以及后续可能的研究趋势进行了分析。     

Indoor scene understanding via RGB-D image segmentation employing depth-based CNN and CRFs

With the availability of low-cost depth-visual sensing devices, such as Microsoft Kinect, we are experiencing a growing interest in indoor environment understanding, at the core of which is semantic segmentation in RGB-D image. The latest research shows that the convolutional neural network (CNN) still dominates the image semantic segmentation field. However, down-sampling operated during the training process of CNNs leads to unclear segmentation boundaries and poor classification accuracy. To address this problem, in this paper, we propose a novel end-to-end deep architecture, termed FuseCRFNet, which seamlessly incorporates a fully-connected Conditional Random Fields (CRFs) model into a depth-based CNN framework. The proposed segmentation method uses the properties of pixel-to-pixel relationships to increase the accuracy of image semantic segmentation. More importantly, we formulate the CRF as one of the layers in FuseCRFNet to refine the coarse segmentation in the forward propagation, in meanwhile, it passes back the errors to facilitate the training. The performance of our FuseCRFNet is evaluated by experimenting with SUN RGB-D dataset, and the results show that the proposed algorithm is superior to existing semantic segmentation algorithms with an improvement in accuracy of at least 2%, further verifying the effectiveness of the algorithm.

     

Face recognition: a convolutional neural-network approach

We present a hybrid neural-network for human face recognition which compares favourably with other methods. The system combines local image sampling, a self-organizing map (SOM) neural network, and a convolutional neural network. The SOM provides a quantization of the image samples into a topological space where inputs that are nearby in the original space are also nearby in the output space, thereby providing dimensionality reduction and invariance to minor changes in the image sample, and the convolutional neural network provides partial invariance to translation, rotation, scale, and deformation. The convolutional network extracts successively larger features in a hierarchical set of layers. We present results using the Karhunen-Loeve transform in place of the SOM, and a multilayer perceptron (MLP) in place of the convolutional network for comparison. We use a database of 400 images of 40 individuals which contains quite a high degree of variability in expression, pose, and facial details. We analyze the computational complexity and discuss how new classes could be added to the trained recognizer.     

基于级联CNNs的非约束车牌精确定位

为解决单一深层卷积神经网络用于非约束场景下车牌定位时,所输出的矩形检测框对非正面车牌定位效果不佳的问题,提出将目标检测与目标分类CNN网络级联,通过检测网络得到感兴趣区域,接着利用轻量级分类网络,将车牌顶点检测问题转化为回归问题。首先,利用YOLOv3网络进行粗定位,获取图像中所有车牌的候选区域;然后,使用基于MobileNetV3改进的轻量级神经网络定位候选区域中的车牌顶点,实现车牌区域精定位;最后,通过透视变换将车牌区域投影到矩形框内实现车牌校正。实验结果表明,所提出的级联CNNs能够有效解决单一CNN目标检测网络仅能输出矩形检测框,而不适用于非约束车牌定位的问题,具有较好的应用价值。