Classification Of Ground Objects Research Articles

Research on classification method based on multi-scale segmentation and hierarchical classification

This paper transmission line corridors covering area in Hebei north area as the research object to explore multi-scale segmentation threshold suitable for Hebei north image, found applicable to Hebei north region segmentation threshold rules. Main methods are the object-oriented multi-scale segmentation and hierarchical classification, using image segmentation principle, make full use of high resolution image rich features such as shape, texture, object relationships. It is used for the follow-up investigation of hidden danger of external damage of power transmission channel in northern Hebei region. The main conclusions of the experiment are as follows: 1. By comparing and analyzing the results of five groups of different thresholds (40, 50, 60, 70 and 80), it is concluded that the single threshold of the multi-scale segmentation method suitable for most mountainous images in the study area is 60, and this method can achieve high-precision ground object classification for images in northern Hebei region. 2. The ground feature cover classification method that is more suitable for the parallel processing of a large number of study areas is the object-oriented hierarchical classification method, and preliminary exploration results of detailed parameters have been achieved.

Hyperspectral Image Classification Based on Capsule Network

The conventional convolutional neural network performs not well enough in the ground objects classification because of its insufficient ability in maintaining sensitive spectral information and characterizing the covariance of spatial structure, resulting from the narrow sensitive frequency band and complex spatial structure with diversity of hyperspectral remote sensing data which caused more serious phenomena of “same material, different spectra” and “different material, same spectra”. Therefore, an improved capsule network is proposed and introduced into hyperspectral image target recognition. A convolution structure combining shallow features and multi-scale depth features is put forward to reduce the phenomena of “different material, same spectra” firstly, and then the diversity of the spatial structure is expressed by the capsule vector and sub-capsule division in channel wise, so that the averaging effect of the convolution process is weakened in the spectral domain and the spatial domain to reduce the phenomena of “same material, different spectra”. By comparing the experimental results on the hyperspectral data sets such as Indian Pines, Salinas, Tea Tree and Xiongan, the capsule network shows strong spatial structure expression ability, flexible deep and shallow feature fusion ability in multi-scale, and its accuracy in target recognition is better than that of conventional convolutional neural networks, so it is suitable for the recognition of complex targets in hyperspectral images.

Efficient Convolutional Neural Architecture Search for LiDAR DSM Classification

Light detection and ranging (LiDAR) data provides rich elevation information, so it plays an irreplaceable role in ground objects classification. Recently, convolutional neural networks (CNNs) have shown excellent performance in LiDAR Digital Surface Models (DSM) classification. However, the architecture of CNN model relies heavily on manual design, so it has great limitations. In addition, different sensors capture LiDAR datasets with different properties, so the model should be designed to suit for different datasets, which further increases the workload of architecture design. Therefore, this paper proposes a method of automatic design of LiDAR DSM classification model. First, attention mechanism is introduced into search space to improve the feature extraction capability of the model. Then a gradient-based search strategy is used to obtain the optimal architecture from this search space. Second, a learning rate adjustment strategy is proposed to reduce the time spent in the search stage and evaluation stage to improve the classification accuracy of the model. Finally, a regularization scheme is introduced to enhance the robustness of the model and avoid overfitting. Experimental results on three public LiDAR datasets (Bayview Park, Recology and Houston) obtained from different sensors show that the proposed neural architecture search method achieves the impressive classification performance compared to several state-of-the-art classification methods and improves the classification accuracy under the condition of limited training samples.

Polarized multispectral image classification of typical ground objects based on neural network (Invited)

Polarized multispectral image classification of typical ground objects based on neural network (Invited)

Hyperspectral Remote Sensing Image Feature Representation Method Based on CAE-H with Nuclear Norm Constraint

Due to the high dimensionality and high data redundancy of hyperspectral remote sensing images, it is difficult to maintain the nonlinear structural relationship in the dimensionality reduction representation of hyperspectral data. In this paper, a feature representation method based on high order contractive auto-encoder with nuclear norm constraint (CAE-HNC) is proposed. By introducing Jacobian matrix in the CAE of the nuclear norm constraint, the nuclear norm has better sparsity than the Frobenius norm and can better describe the local low dimension of the data manifold. At the same time, a second-order penalty term is added, which is the Frobenius norm of the Hessian matrix expressed in the hidden layer of the input, encouraging a smoother low-dimensional manifold geometry of the data. The experiment of hyperspectral remote sensing image shows that CAE-HNC proposed in this paper is a compact and robust feature representation method, which provides effective help for the ground object classification and target recognition of hyperspectral remote sensing image.

Study on ground object classification based on the hyperspectral fusion images of ZY-1(02D) satellite

Spaceborne hyperspectral images are expected to have abundant applications in various fields, particularly in the quantitative observation of the Earth. However, the problem of low spatial resolution has limited their effectiveness to some extent. There are urgent needs for high-spatial-resolution hyperspectral images. We explore fusing the hyperspectral image with panchromatic and multispectral images captured by sensors onboard ZY-1 (02D) to obtain high-spatial-resolution and high-spectral-resolution images. Four approaches to obtain the final high-spatial-resolution hyperspectral image data are proposed, and six fusion methods are used. The fusion results were evaluated using quantitative indicators and classification performance, and image quality greatly improved after fusion. The fusion approach that used multispectral image data as an intermediate layer twice in the fusion process obtained the best fusion effect, and this was verified by both quantitative and application evaluation results. This provides an effective approach to obtain a high-quality, high-spatial-resolution hyperspectral image using the combination of panchromatic, multispectral, and hyperspectral images acquired by the sensors onboard ZY-1(02D).

Research on Estimation of Paddy Field Area Index Based on UAV Remote Sensing Images

In order to verify the superiority and effectiveness of extracting rice information based on UAV images. This paper takes the rice plot as the research object, and uses the portable UAV Mavic Pro for aerial photography. Preprocess the acquired UAV images to generate orthophotos with a resolution of 3.95cm/pix. Using object-oriented thinking, visual evaluation and ESP tools are combined to quickly select the optimal segmentation scale to be 300, and support is applied. Vector machine, random forest, and nearest neighbor supervised classification methods have carried out ground object classification and rapid extraction of rice area. The classification results and area accuracy are evaluated by visual classification results. The method with the highest overall accuracy is the nearest neighbor classification method. At this time, the user accuracy of rice classification is 95%, and the area consistency accuracy is 99%. The results show that UAV remote sensing and automatic classification can quickly obtain high resolution images and extract rice planting area in plain rice planting area, make up for the lack of ground survey data when Nongshan is blocked, and provide samples and verification basis for the calculation of large-scale rice planting area, yield and other information.

Spectral-Spatial Hyperspectral Image Semisupervised Classification by Fusing Maximum Noise Fraction and Adaptive Random Multigraphs

Hyperspectral images (HSIs) contain large amounts of spectral and spatial information, and this provides the possibility for ground object classification. However, when using the traditional method, achieving a satisfactory classification result is difficult because of the insufficient labeling of samples in the training set. In addition, parameter adjustment during HSI classification is time-consuming. This paper proposes a novel fusion method based on the maximum noise fraction (MNF) and adaptive random multigraphs for HSI classification. Considering the overall spectrum of the object and the correlation of adjacent bands, the MNF was utilized to reduce the spectral dimension. Next, a multiscale local binary pattern (LBP) analysis was performed on the MNF dimension-reduced data to extract the spatial features of different scales. The obtained multiscale spatial features were then stacked with the MNF dimension-reduced spectral features to form multiscale spectral-spatial features (SSFs), which were sent into the RMG for HSI classification. Optimal performance was obtained by fusion. For all three real datasets, our method achieved competitive results with only 10 training samples. More importantly, the classification parameters corresponding to different hyperspectral data can be automatically optimized using our method.

3D-CNN based UAV hyperspectral imagery for grassland degradation indicator ground object classification research

The identification and counting of grassland degradation indicator ground objects is an important component of grassland ecological monitoring. These steps are also an important basis for developing ecological restoration and management programs for degraded grasslands. Compared with a traditional human survey, the use of remote sensing images can not only achieve dynamic monitoring of a large area, but also improve the efficiency. Recently, most studies regarding ground object classification based on remote sensing images address the development and optimization of classification models for features in several widely used datasets. For the remote sensing of desertified grasslands, remote sensing images with high spatial resolutions are used for studies on small and sparse features in degraded grasslands. The spatial resolution of the above mentioned datasets yields difficulties when attempting to classify small and sparse indicator features for desertified grasslands because generalization becomes limited. Therefore, establishing a lightweight classification model suitable for degraded grassland features with high spatial resolution is important. In this study, a low altitude unmanned aerial vehicle (UAV) hyperspectral remote sensing platform was constructed to collect high spatial resolution remote sensing images of degraded grasslands. The GDIF-3D-CNN classification model was used to classify the pure pixels and all pixels datasets, whose accuracy and efficiency were further improved by optimizing the eight parameters of the model. This study explores the remote sensing ground object classification of thin small plants and a large number of mixed pixels, realizing high precision classification among desertification degradation indicating plant populations of a species, and provides key quantitative data for grassland degradation research.

Hyperspectral Image Classification Based on Stacked Contractive Autoencoder Combined With Adaptive Spectral-Spatial Information

Hyperspectral image (HSI) contain abundant spectral and spatial information, enabling the accurate classification of ground objects. However, many existing machine learning methods have poor performance, and some existing CNN-based methods require high computational power, which considerably limits their real-world applications. To address these issues, in this paper, we propose an alternative HSI classification method based on the stacked contractive autoencoder (SCAE) and adaptive spectral-spatial information to improve the accuracy of HSI classification. Specifically, the non-subsampled shearlet transform (NSST) with the guided filtering (NG) enhances spatial structure information. Subsequently, we present an adaptive spatial information extraction method to extract the spatial information of pixels. Furthermore, we propose an HSI classification network, called SCAE-LR, for feature extraction and classification. The SCAE is implemented to extract the adaptive spectral-spatial feature, and a logistic regression (LR) layer is employed for classification. Extensive experiments on the Indian Pines data set and the Pavia University data set demonstrate the superior performance of our method.

Hyperspectral Image Classification Based on Multilevel Joint Feature Extraction Network

Over the past few years, convolutional neural network (CNN) has been broadly adopted in remote sensing (RS) imagery processing areas due to its impressive capabilities in feature extraction. Nevertheless, it is still a challenge for CNN-based hyperspectral image (HSI) classification methods to extract more effective spectral-spatial features considering all spectral bands. Driven by this issue, we propose a novel approach to cope with the HSI classification task, referring to the multilevel joint feature extraction network. The proposed network makes full use of the information on each channel of HSI and transforms it into valid channel-wised spatial features through a designed convolution process. Moreover, these feature maps form global attention details to guide the extraction of spectral-spatial features, which are taken to the next level for further feature mining. Then, the features obtained at different levels are integrated for ground object classification. In contrast with several state-of-the-art HSI classification methods on four public datasets, experimental results demonstrate the effectiveness and remarkable feature extraction capability of our proposed approach.

Establishment and application of the modified Chen NDVI model integrated with ground object classification

Heterogeneous land surface causes the scale effect of remotely sensed land surface parameters. Addressing on quantitatively describing the influence of different ground objects on scale effect of the common surface parameter normalized difference vegetation index (NDVI), the paper proposed an improved NDVI scale transformation model. The model integrated accurate classification information from medium- or high- spatial resolution remote sensing images to improve the traditional Chen NDVI scale conversion model, and showed its superiority for NDVI scale effect description. Xiamen was taken as the experimental area for the study and the conclusions could be obtained from the experimental results. Compared with the traditional Chen NDVI model with rough information, the improved Chen NDVI model incorporating fine ground information provides a finer and more quantitative description of the influence of different land types on the NDVI scale effect. Furthermore, it is found that the presence of water is the key factor underlying the NDVI scale effect. The conclusions of this study have important implications for the scale effect research of other NDVI-like surface parameters such as ratio vegetation index (RVI), normalized difference built-up index (NDBI), normalized burn ratio (NBR).

An Overview on Linear Unmixing of Hyperspectral Data

Hyperspectral remote sensing technology has a strong capability for ground object detection due to the low spatial resolution of hyperspectral imaging spectrometers. A single pixel that leads to a hyperspectral remote sensing image usually contains more than one feature coverage type, resulting in a mixed pixel. The existence of a mixed pixel affects the accuracy of the ground object identification and classification and hinders the application and development of hyperspectral technology. For the problem of unmixing of mixed pixels in hyperspectral images (HSIs), the linear mixing model can model the mixed pixels well. Through the collation of nearly five years of the literature, this paper introduces the development status and problems of linear unmixing models from four aspects: geometric method, nonnegative matrix factorization (NMF), Bayesian method, and sparse unmixing.

Fusion of Hyperspectral CASI and Airborne LiDAR Data for Ground Object Classification through Residual Network.

Modern satellite and aerial imagery outcomes exhibit increasingly complex types of ground objects with continuous developments and changes in land resources. Single remote-sensing modality is not sufficient for the accurate and satisfactory extraction and classification of ground objects. Hyperspectral imaging has been widely used in the classification of ground objects because of its high resolution, multiple bands, and abundant spatial and spectral information. Moreover, the airborne light detection and ranging (LiDAR) point-cloud data contains unique high-precision three-dimensional (3D) spatial information, which can enrich ground object classifiers with height features that hyperspectral images do not have. Therefore, the fusion of hyperspectral image data with airborne LiDAR point-cloud data is an effective approach for ground object classification. In this paper, the effectiveness of such a fusion scheme is investigated and confirmed on an observation area in the middle parts of the Heihe River in China. By combining the characteristics of hyperspectral compact airborne spectrographic imager (CASI) data and airborne LiDAR data, we extracted a variety of features for data fusion and ground object classification. Firstly, we used the minimum noise fraction transform to reduce the dimensionality of hyperspectral CASI images. Then, spatio-spectral and textural features of these images were extracted based on the normalized vegetation index and the gray-level co-occurrence matrices. Further, canopy height features were extracted from airborne LiDAR data. Finally, a hierarchical fusion scheme was applied to the hyperspectral CASI and airborne LiDAR features, and the fused features were used to train a residual network for high-accuracy ground object classification. The experimental results showed that the overall classification accuracy was based on the proposed hierarchical-fusion multiscale dilated residual network (M-DRN), which reached an accuracy of 97.89%. This result was found to be 10.13% and 5.68% higher than those of the convolutional neural network (CNN) and the dilated residual network (DRN), respectively. Spatio-spectral and textural features of hyperspectral CASI images can complement the canopy height features of airborne LiDAR data. These complementary features can provide richer and more accurate information than individual features for ground object classification and can thus outperform features based on a single remote-sensing modality.

A modified Cuckoo Search algorithm based optimal band subset selection approach for hyperspectral image classification

Band selection is an effective way to reduce the size of hyperspectral data and to overcome the “curse of dimensionality” in ground object classification. This paper presents a band selection approach based on modified Cuckoo Search (CS) optimisation with correlation-based initialisation. CS is a popular metaheuristic algorithm with efficient optimisation capabilities for band selection. However, it can easily fall into local optimum solutions. To avoid falling into a local optimum, an initialisation strategy based on correlation is adopted instead of random initialisation to initiate the location of nests. Experimental results with Indian Pines, Salinas and Pavia University datasets show that the proposed approach obtains overall accuracy of 82.83 %, 94.83 % and 91.79 %, respectively, which is higher than the original CS algorithm, Genetic Algorithm (GA), Particle Swarm Optimisation (PSO) and Gray Wolf Optimisation (GWO).

A survey of band selection techniques for hyperspectral image classification

Hyperspectral images usually contain hundreds of contiguous spectral bands, which can precisely discriminate the various spectrally similar classes. However, such high-dimensional data also contain highly correlated and irrelevant information, leading to the curse of dimensionality (also called the Hughes phenomenon). It is necessary to reduce these bands before further analysis, such as land cover classification and target detection. Band selection is an effective way to reduce the size of hyperspectral data and to overcome the curse of the dimensionality problem in ground object classification. Focusing on the classification task, this article provides an extensive and comprehensive survey on band selection techniques describing the categorisation of methods, methodology used, different searching approaches and various technical difficulties, as well as their performances. Our purpose is to highlight the progress attained in band selection techniques for hyperspectral image classification and to identify possible avenues for future work, in order to achieve better performance in real-time operation.

A RADIOMETRIC NORMALIZATION METHOD OF CONTROLLING NO-CHANGED SET (CNCS) FOR DIVERSE LANDCOVER USING MULTI-SENSOR DATA

Abstract. The accurate acquisition of land surface reflectance (SR) data determines the accuracy of ground objects recognition, classification and land surface parameter inversion using remote sensing data, which is the basis of remote sensing data application. In this study, a Control No-Changed Set (CNCS) radiometric normalization method is proposed to realize spectral information transformation of multi-sensor data, which is based on the Iteratively Reweighted Multivariate Alteration Detection (IR-MAD), and includes automatic selection and step-by-step optimization of no-change pixels. The No-Changed set (NC) is obtained by selecting the original no-change pixels between the target image and the reference image according to the linear relationship. In the obtained original no-change regions, IR-MAD rules with iterative control are used to fix the final no-change pixels, after regression modeling and calculation, the normalized images are obtained. The method is tested on multi-images from multi-sensors in three groups of experiments (GF-1 WFV and Landsat-8 OLI, GF-1 PMS and Sentinel-2 MSI, and Landsat-8 OLI and Sentinel-2 MSI) with different landcover areas. The results of radiometric normalization are evaluated qualitatively and quantitatively. The data of the three groups of experiments have a high correlation (correlation coefficient r values > 0.85), indicating that they can be used together as complementary data. The Root Mean Squared Error (RMSE) values calculate from the NC between the reference and normalized target images are much smaller than those between the reference and original target images. The radiometric colour composition effects, and the typical ground objects spectral reflective curves of the reference and normalized target images are very similar after radiometric normalization. These results indicate that the CNCS method considers the linear relationship of the no-change pixels and is effective, stable, and can be used to improve the consistency of SR of multi-images from multi-sensors.

Virtual design of urban planning based on GIS big data and machine learning

In order to build a virtual urban planning model and improve the effect of urban planning, this paper builds a virtual urban planning design model based on GIS big data technology and machine learning algorithms, and proposes a solution that combines multiple features. With the development of polarized SAR in the direction of high resolution, a single feature often cannot fully express the detailed information of ground objects, resulting in poor classification results and low accuracy. The combination of multiple features can express feature information well. In addition, this paper uses the ELM method to plan SAR ground object classification, uses an extreme learning machine classification algorithm with fast learning speed and good classification effect, and uses ELM as a classifier. Finally, this paper designs experiments to explore the performance of the model constructed in this paper from two aspects: detection accuracy and planning score. The research results show that the model constructed in this paper meets the expected goals.

Deep Belief Network for Spectral⁻Spatial Classification of Hyperspectral Remote Sensor Data.

With the development of high-resolution optical sensors, the classification of ground objects combined with multivariate optical sensors is a hot topic at present. Deep learning methods, such as convolutional neural networks, are applied to feature extraction and classification. In this work, a novel deep belief network (DBN) hyperspectral image classification method based on multivariate optical sensors and stacked by restricted Boltzmann machines is proposed. We introduced the DBN framework to classify spatial hyperspectral sensor data on the basis of DBN. Then, the improved method (combination of spectral and spatial information) was verified. After unsupervised pretraining and supervised fine-tuning, the DBN model could successfully learn features. Additionally, we added a logistic regression layer that could classify the hyperspectral images. Moreover, the proposed training method, which fuses spectral and spatial information, was tested over the Indian Pines and Pavia University datasets. The advantages of this method over traditional methods are as follows: (1) the network has deep structure and the ability of feature extraction is stronger than traditional classifiers; (2) experimental results indicate that our method outperforms traditional classification and other deep learning approaches.

Методика автоматизированного создания топографического плана на основе съёмки с беспилотного летательного аппарата

The authors show an automated method for performing the main stages of creating a topographic plan. The optimal number and location of the reference points for creation a digital terrain model using drones is determined. The necessary components of a multi-rotor helicopter are described. They are required for lifting the camera into the air and increasing the duration of the flight. On the basis of the research, a significant speed increase of the field work was achieved using effective satellite and linear-angular measurements for determination of the reference points’ coordinates and productive survey from anunmanned aerial vehicle. Algorithms forconstructing an orthophoto and a digital terrain model as well as automated filtering of the resulting dense point cloud for creating a digital surface model are presented. The high-accurate modification of the OBIA (Object-Based Image Analysis) algorithm for classification of ground objects is determined. At the end of the article, the algorithm for automated vectorization of the raster classification using the ArcGIS geoinformation software and converting of the received objects to convention for creating an electronic topographic plan is given.