High-resolution Remote Sensing Image Classification Research Articles

Object-Based Semi-Supervised Spatial Attention Residual UNet for Urban High-Resolution Remote Sensing Image Classification

Object-Based Semi-Supervised Spatial Attention Residual UNet for Urban High-Resolution Remote Sensing Image Classification

Joint superpixel and Transformer for high resolution remote sensing image classification

Deep neural networks combined with superpixel segmentation have proven to be superior to high-resolution remote sensing image (HRI) classification. Currently, most HRI classification methods that combine deep learning and superpixel segmentation use stacking on multiple scales to extract contextual information from segmented objects. However, this approach does not take into account the contextual dependencies between each segmented object. To solve this problem, a joint superpixel and Transformer (JST) framework is proposed for HRI classification. In JST, HRI is first segmented into superpixel objects as input, and Transformer is used to model the long-range dependencies. The contextual relationship between each input superpixel object is obtained and the class of analyzed objects is output by designing an encoding and decoding Transformer. Additionally, we explore the effect of semantic range on classification accuracy. JST is also tested by using two HRI datasets with overall classification accuracy, average accuracy and Kappa coefficients of 0.79, 0.70, 0.78 and 0.91, 0.85, 0.89, respectively. The effectiveness of the proposed method is compared qualitatively and quantitatively, and the results achieve competitive and consistently better than the benchmark comparison method.

Change Detection Trend Prediction of Leisure Agriculture and Rural Tourism Based on High-Resolution Remote Sensing Image Classification Algorithm

Change Detection Trend Prediction of Leisure Agriculture and Rural Tourism Based on High-Resolution Remote Sensing Image Classification Algorithm

A decoupled search deep network framework for high-resolution remote sensing image classification

ABSTRACT Currently, high-resolution remote sensing image (HRI) interpretation is often done using deep learning methods. However, most deep learning networks currently used to classify HRIs rely on manually created skeletons, which are labour- and time-intensive and unable to completely identify the intrinsic features of the target image. To solve these problems, this study proposes a deep network skeleton search method based on a three-layer search space. A decoupled search approach was designed to optimize this three-layer search space. Our results demonstrated that the network architecture based on the decoupled search of the depth network skeleton outperforms the manually created depth network currently used for HRI classification. In particular, the mean Intersection over Union (mIoU), Accuracy (Acc) and Frequency Weighted Intersection over Union (FWIoU) values of the proposed method on the first test dataset were 0.902, 0.882 and 0.913, respectively, and those of the second test dataset were 0.887, 0.893 and 0.911, respectively. This method enables the development of autonomous designs of network skeletons, namely for HRI feature extraction and classification.

DMAU-Net: An Attention-Based Multiscale Max-Pooling Dense Network for the Semantic Segmentation in VHR Remote-Sensing Images

High-resolution remote-sensing images cover more feature information, including texture, structure, shape, and other geometric details, while the relationships among target features are more complex. These factors make it more complicated for classical convolutional neural networks to obtain ideal results when performing a feature classification on remote-sensing images. To address this issue, we proposed an attention-based multiscale max-pooling dense network (DMAU-Net), which is based on U-Net for ground object classification. The network is designed with an integrated max-pooling module that incorporates dense connections in the encoder part to enhance the quality of the feature map, and thus improve the feature-extraction capability of the network. Equally, in the decoding, we introduce the Efficient Channel Attention (ECA) module, which can strengthen the effective features and suppress the irrelevant information. To validate the ground object classification performance of the multi-pooling integration network proposed in this paper, we conducted experiments on the Vaihingen and Potsdam datasets provided by the International Society for Photogrammetry and Remote Sensing (ISPRS). We compared DMAU-Net with other mainstream semantic segmentation models. The experimental results show that the DMAU-Net proposed in this paper effectively improves the accuracy of the feature classification of high-resolution remote-sensing images. The feature boundaries obtained by DMAU-Net are clear and regionally complete, enhancing the ability to optimize the edges of features.

Novel Knowledge Graph- and Knowledge Reasoning-Based Classification Prototype for OBIA Using High Resolution Remote Sensing Imagery

Although many machine learning methods have been successfully applied for the object-based classification of high resolution (HR) remote sensing imagery, current methods are highly dependent on the spectral similarity between segmented objects and have disappointingly poor performance when dealing with different segmented objects that have similar spectra. To overcome this limitation, this study exploited a knowledge graph (KG) that preserved the spatial relationships between segmented objects and has a reasoning capability that can assist in improving the probability of correctly classifying different segmented objects with similar spectra. In addition, to assist the knowledge graph classifications, an image segmentation method generating segmented objects that closely resemble real ground objects in size was used, which improves the integrity of the object classification results. Therefore, a novel HR remote sensing image classification scheme is proposed that involves a knowledge graph and an optimal segmentation algorithm, which takes full advantage of object-based classification and knowledge inference. This method effectively addresses the problems of object classification integrity and misclassification of objects with the same spectrum. In the evaluation experiments, three QuickBird-2 images and over 15 different land cover classes were utilized. The results showed that the classification accuracy of the proposed method is high, with overall accuracies exceeding 0.85. These accuracies are higher than the K Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) methods. The evaluated results confirmed that the proposed method offers excellent performance in HR remote sensing image classification.

DASFNet: Dense-Attention–Similarity-Fusion Network for scene classification of dual-modal remote-sensing images

• The cosine similarity and a gating mechanism are used by the SFM to fuse IRRG and DSM data and extract corresponding complementary features. • The MCFAM is used to extract multiscale context cues, solving the problem of object-scale variations in HRRSIs. • The DAMM forms a dense-attention branch that can adaptively capture both low- and high-level attentions. Although significant progress has been made in scene classification of high-resolution remote-sensing images (HRRSIs), dual-modal HRRSI scene classification is still an active and challenging issue. In this study, we introduce an end-to-end dense-attention–similarity-fusion network (DASFNet) for dual-modal HRRSIs. Specifically, we propose a dense-attention map module based on graph convolution, which adaptively captures long-range semantic cues and further directs shallow-attention cues to the deep level to guide the generation of high-level feature attention cues. At the encoder stage, DASFNet uses feature similarity to explore the correlation between dual-modal features; a similarity-fusion module extracts complementary information by fusing features from different modalities. A multiscale context-feature-aggregation module is used to strengthen the feature embedding of any two spatial scales; this solves the of scale change problem. A large number of experiments on two HRRSI benchmark datasets for scene classification indicate that the proposed DASFNet outperforms the outstanding scene classification approaches.

Enhanced shuffle attention network based on visual working mechanism for high-resolution remote sensing image classification

High-resolution remote sensing images provide complete and detailed ground scenes. However, these complex and diverse details make it difficult to interpret large area objects and their small area details at a single granularity. Therefore, the method proposed in this paper refers to the principles of visual working memory, visual attention and visual reasoning, independently analyse the ground objects of different granularities, and integrates the recognition results of the different granularity ground objects to form accurate classification results. This multi-granularity classification strategy suppresses noise interference, protects typical edges, and highlights clear details. Experimental results with dataset of real land use survey projects, including images of different time, multiple perspectives, and different altitudes, The results show that the idea of combining unsupervised edge detection with neural network classification can greatly enhance the generalization ability and edge cognitive ability, making the classification accuracy reach 80% on average, with an improvement of about 5%.

An improved categorical cross entropy for remote sensing image classification based on noisy labels

In recent years, volunteered geographic information (VGI) is widely used to train deep convolutional neural networks (DCNNs) for high-resolution remote sensing image classification. However, noisy labels are often included in the training samples generated by VGI, which inevitably affects the performance of DCNNs. To solve this problem, the negative effect of noisy labels on remote sensing image classification by DCNNs is analyzed. Then, an improved categorical cross-entropy (ICCE) is proposed to address the issue of noisy labels. The ICCE improves the robustness of DCNN to noisy labels by revisiting the sample weighting scheme so that much attention is paid to the clean samples instead of the noisy samples. Besides, the error bound of ICCE is derived with strict mathematical proof under different types of noisy labels, which ensures the advantages of ICCE from theory. Additionally, extensive experiments are conducted on three remote sensing image datasets with simulated and real noisy labels to quantitatively evaluate the performance of ICCE.

An Object-Oriented CNN Model Based on Improved Superpixel Segmentation for High-Resolution Remote Sensing Image Classification

Object-oriented convolutional neural network (OCNN) has been proven to be an effective classification method for very fine spatial resolution (VFSR) remotely sensed imagery. It can obtain higher accuracy and well edge preservation results due to the combination of advantages of image segmentation and deep network at the same time. However, the mismatch with the real boundary of the ground object is still a problem that needs to be solved further. In addition, a specific CNN model that can learn better feature representations also plays an important role in improving classification accuracy. For these purposes, we proposed an improved sample linear iterative cluster (ISLIC) to obtain better segmentation edges. This algorithm overcomes the limitation of the input feature dimension in SLIC and improves the boundary performance by using more features. Besides, in order to obtain better feature representations, a new CNN model has also been developed, which can make full use of spectral information to learn first-order and second-order fusion features for classification. This method has been verified on four real remote sensing images. Compared with other methods, the proposed method achieves better performance in terms of edge and classification accuracy.

Landslide Extraction from High-Resolution Remote Sensing Imagery Using Fully Convolutional Spectral–Topographic Fusion Network

Considering the complexity of landslide hazards, their manual investigation lacks efficiency and is time-consuming, especially in high-altitude plateau areas. Therefore, extracting landslide information using remote sensing technology has great advantages. In this study, comprehensive research was carried out on the landslide features of high-resolution remote sensing images on the Mangkam dataset. Based on the idea of feature-driven classification, the landslide extraction model of a fully convolutional spectral–topographic fusion network (FSTF-Net) based on a deep convolutional neural network of multi-source data fusion is proposed, which takes into account the topographic factor (slope and aspect) and the normalized difference vegetation index (NDVI) as multi-source data input by which to train the model. In this paper, a high-resolution remote sensing image classification method based on a fully convolutional network was used to extract the landslide information, thereby realizing the accurate extraction of the landslide and surrounding ground-object information. With Mangkam County in the southeast of the Qinghai–Tibet Plateau China as the study area, the proposed method was evaluated based on the high-precision digital elevation model (DEM) generated from stereoscopic images of Resources Satellite-3 and multi-source high-resolution remote sensing image data (Beijing-2, Worldview-3, and SuperView-1). Results show that our method had a landslide detection precision of 0.85 and an overall classification accuracy of 0.89. Compared with the latest DeepLab_v3+, our model increases the landslide detection precision by 5%. Thus, the proposed FSTF-Net model has high reliability and robustness.

High-resolution remote sensing image classification based on improved DeepLabV3 network

High-resolution remote sensing image classification based on improved DeepLabV3 network

High-Resolution Remote Sensing Image Classification with RmRMR-Enhanced Bag of Visual Words

A ReliefF improved mRMR (RmRMR) criterion-based bag of visual words (BoVW) algorithm is proposed to filter the visual words that are generated with high information redundancy for remote sensing image classification. First, the contribution degree of each word to the classification is represented by its weighting parameter, which is assigned using the ReliefF algorithm. Next, the relevance and redundancy of each word are calculated according to the mRMR criterion with the addition of a dictionary balance coefficient. Finally, a novel dictionary discriminant function is established, and the globally discriminative small-scale dictionary subsets are filtered and obtained. Experimental results show that the proposed algorithm effectively reduces the amount of redundant information in the dictionary and better balances the relevance and redundancy of words to improve the feature descriptive power of dictionary subsets and markedly increase the classification precision on a high-resolution remote sensing image.

Land-cover classification of high-resolution remote sensing image based on multi-classifier fusion and the improved Dempster–Shafer evidence theory

The use of the single machine learning classifier for high-resolution remote sensing (RS) image classification makes it difficult to improve the accuracy of classification results. To fully utilize the advantages of different classifiers for different types of ground objects, based on the Dempster–Shafer (DS) evidence theory, we propose a multi-classifier fusion method for classification of high-resolution RS images. Six machine learning classifiers: support vector machines, k-nearest neighbor, random forest, artificial neural network, classification and regression tree, and the C5.0 decision tree were selected for application in the fusion of multiple classifiers. We calculated a classifier difference index based on the accuracy and difference of the classification results of the base classifiers. Base classifiers with large differences were selected to perform integration based on the DS evidence theory. We also improved the classical DS evidence theory. First, based on the classification validity of the base classifier for different ground objects, the classification probability value of the base classifier for different samples was weight optimized. Then different fusion methods were selected according to the classification conflict coefficients between base classifiers. The results reveal that the overall accuracy and kappa coefficients of the fusion classifier are significantly better than those of the base classifier. The producer’s accuracy and user’s accuracy of the fusion results based on the improved DS evidence theory were higher than those of the fusion results based on the classical DS evidence theory.

Standardized object-based dual CNNs for very high-resolution remote sensing image classification and standardization combination effect analysis

ABSTRACT Advances in the object-based convolutional neural network (CNN) have demonstrated the superiority of CNNs for image classification. However, any object-based CNN, regardless of its model structure, only stacks the square images with different scales when extracting features. The impact of background information around the segmented object (the number of pixels around the segmented object) for the classification accuracy is neglected. In addition, blurred object boundaries and feature representation, as well as huge computational redundancy, restrict the application for very high-resolution remote sensing image (VHRI) classification. To solve these problems, a novel standardized object-based dual CNN (SOD-CNN) is proposed for VHRI classification. First, based on geographic object-based image analysis, the image is segmented into homogeneous regions. Second, these less-segmented objects are over-segmented into superpixels with high compactness to provide crisp and accurate boundary delineation at the pixel level. Third, four standardization methods are developed to limit the number of pixels around the segmented object. The standardized less-segmented object and over-segmented object are fed into two different CNNs to capture different perspectives of features at local and global scales. Finally, feature fusion based on the full connection is performed to integrate the class-specific classification results. The effectiveness of the proposed method was verified by using two VHRI, which achieved excellent classification accuracy, consistently outperforming the benchmark comparisons. The overall and per-class classification accuracy was investigated under different standardization combinations. We found that (1) the proposed standardization method not only reduced redundancy of information in the object-based CNN but also highlighted the features of segmented objects; (2) different segmented objects had different optimal standardization combinations; and (3) the classification accuracy was reasonably controlled by the foundation number of training samples.

Hierarchical Multi-View Semi-Supervised Learning for Very High-Resolution Remote Sensing Image Classification

Traditional classification methods used for very high-resolution (VHR) remote sensing images require a large number of labeled samples to obtain higher classification accuracy. Labeled samples are difficult to obtain and costly. Therefore, semi-supervised learning becomes an effective paradigm that combines the labeled and unlabeled samples for classification. In semi-supervised learning, the key issue is to enlarge the training set by selecting highly-reliable unlabeled samples. Observing the samples from multiple views is helpful to improving the accuracy of label prediction for unlabeled samples. Hence, the reasonable view partition is very important for improving the classification performance. In this paper, a hierarchical multi-view semi-supervised learning framework with CNNs (HMVSSL) is proposed for VHR remote sensing image classification. Firstly, a superpixel-based sample enlargement method is proposed to increase the number of training samples in each view. Secondly, a view partition method is designed to partition the training set into two independent views, and the partitioned subsets are characterized by being inter-distinctive and intra-compact. Finally, a collaborative classification strategy is proposed for the final classification. Experiments are conducted on three VHR remote sensing images, and the results show that the proposed method performs better than several state-of-the-art methods.

An End-to-End and Localized Post-Processing Method for Correcting High-Resolution Remote Sensing Classification Result Images

Since the result images obtained by deep semantic segmentation neural networks are usually not perfect, especially at object borders, the conditional random field (CRF) method is frequently utilized in the result post-processing stage to obtain the corrected classification result image. The CRF method has achieved many successes in the field of computer vision, but when it is applied to remote sensing images, overcorrection phenomena may occur. This paper proposes an end-to-end and localized post-processing method (ELP) to correct the result images of high-resolution remote sensing image classification methods. ELP has two advantages. (1) End-to-end evaluation: ELP can identify which locations of the result image are highly suspected of having errors without requiring samples. This characteristic allows ELP to be adapted to an end-to-end classification process. (2) Localization: Based on the suspect areas, ELP limits the CRF analysis and update area to a small range and controls the iteration termination condition. This characteristic avoids the overcorrections caused by the global processing of the CRF. In the experiments, ELP is used to correct the classification results obtained by various deep semantic segmentation neural networks. Compared with traditional methods, the proposed method more effectively corrects the classification result and improves classification accuracy.

Semantic segmentation of very high-resolution remote sensing image based on multiple band combinations and patchwise scene analysis

Large intraclass variance and low interclass variance are among the most challenging problems in very high-resolution (VHR) image classification. Semantic segmentation constructed in a deep convolution neural network is used as a classification algorithm conducted via end-to-end training, which combines spectral–spatial features and context information. However, large-scale remote sensing images cannot be directly processed because they are limited by GPU memory and segmentation algorithm. At the same time, classification using single band combinations is also unsatisfactory due to the extraordinary complex features of VHR images. Therefore, a method is proposed based on multiple band combinations and patchwise scene analysis. A complex remote sensing image can be considered as into a combination of simple scenes from multiple patchwise images. And optimal band combinations of each patchwise image are selected according to their scene. The segmentation results of each patchwise image are merged to get the desired results according to geographical coordinates. Our method is validated on the ISPRS 2-D Semantic Labeling dataset of Potsdam, on which results competitive with the state-of-the-art are obtained. The proposed scheme has strong universality and can be used for large-scale high-resolution remote sensing image classification.

Classification of High-Resolution Remote Sensing Images in the Feilaixia Reservoir Based on a Fully Convolutional Network

In technologically underdeveloped areas, water pollution threatens the living environment of local residents, so remote sensing monitoring of the features around reservoirs is necessary. Fully convolutional networks (FCNs) offer great potential for extracting high-resolution features due to their unlimited input image size and higher accuracy compared to convolutional neural networks. Therefore, a proposal to classify WorldView-2 images is implemented with a sixty-eight thousand iterations of fine-tuning and fully trained combined training method based on a fully convolutional network (SEFCN). The chosen images depict the urban area of Yingde, which is located to the northeast of the Feilaixia Reservoir, Qingyuan, Guangdong Province, China. The SEFCN combines an FCN-32s and FCN-16s to better integrate the deep features and shallow features and effectively improve the classification accuracy. Additionally, the loss value fully converges with enough iterations, and the overfitting caused by the superposition of two full trainings is avoided. The SEFCN model achieves the highest accuracy among all compared classification models and obtains the best classified results on the WorldView-2 images, as demonstrated by attaining the highest F1 scores in each category. After the classified images are optimized using a conditional random field, the current status of the study area is analyzed, and several suggestions for land use in the urban area of Yingde are made. The experiments still have deficiencies in the application of high-resolution remote sensing image classification with different sensors, and the classification results can be optimized and improved in other aspects.

An Improved Hybrid Segmentation Method for Remote Sensing Images

Image segmentation technology, which can be used to completely partition a remote sensing image into non-overlapping regions in the image space, plays an indispensable role in high-resolution remote sensing image classification. Recently, the segmentation methods that combine segmenting with merging have attracted researchers’ attention. However, the existing methods ignore the fact that the same parameters must be applied to every segmented geo-object, and fail to consider the homogeneity between adjacent geo-objects. This paper develops an improved remote sensing image segmentation method to overcome this limitation. The proposed method is a hybrid method (split-and-merge). First, a watershed algorithm based on pre-processing is used to split the image to form initial segments. Second, the fast lambda-schedule algorithm based on a common boundary length penalty is used to merge the initial segments to obtain the final segmentation. For this experiment, we used GF-1 images with three spatial resolutions: 2 m, 8 m and 16 m. Six different test areas were chosen from the GF-1 images to demonstrate the effectiveness of the improved method, and the objective function (F (v, I)), intrasegment variance (v) and Moran’s index were used to evaluate the segmentation accuracy. The validation results indicated that the improved segmentation method produced satisfactory segmentation results for GF-1 images (average F (v, I) = 0.1064, v = 0.0428 and I = 0.17).