Extensive identification of landslide boundaries using remote sensing images and deep learning method

The frequent occurrence of extreme weather events has rendered numerous landslides to a global natural disaster issue.It is crucial to rapidly and accurately determine the boundaries of landslides for geohazards evaluation and emergency response.Therefore,the Skip Connection DeepLab neural network(SCDnn),a deep learning model based on 770 optical remote sensing images of landslide,is proposed to improve the accuracy of landslide boundary detection.The SCDnn model is optimized for the over-segmentation issue which occurs in conventional deep learning models when there is a significant degree of similarity between topographical geomorphic features.SCDnn exhibits notable improvements in landslide feature extraction and semantic segmentation by combining an enhanced Atrous Spatial Pyramid Convolutional Block(ASPC)with a coding structure that reduces model complexity.The experimental results demonstrate that SCDnn can identify landslide boundaries in 119 images with MIoU values between 0.8and 0.9;while 52 images with MIoU values exceeding 0.9,which exceeds the identification accuracy of existing techniques.This work can offer a novel technique for the automatic extensive identification of landslide boundaries in remote sensing images in addition to establishing the groundwork for future inve stigations and applications in related domains.

An Intelligent Detection Method for Optical Remote Sensing Images Based on Improved YOLOv7

To address the issue of imbalanced detection performance and detection speed in current mainstream object detection algorithms for optical remote sensing images,this paper proposes a multi-scale object detection model for remote sensing images on complex backgrounds,called DI-YOLO,based on You Only Look Once v7-tiny(YOLOv7-tiny).Firstly,to enhance the model’s ability to capture irregular-shaped objects and deformation features,as well as to extract high-level semantic information,deformable convolutions are used to replace standard convolutions in the original model.Secondly,a Content Coordination Attention Feature Pyramid Network(CCA-FPN)structure is designed to replace the Neck part of the original model,which can further perceive relationships between different pixels,reduce feature loss in remote sensing images,and improve the overall model’s ability to detect multi-scale objects.Thirdly,an Implicitly Efficient Decoupled Head(IEDH)is proposed to increase the model’s flexibility,making it more adaptable to complex detection tasks in various scenarios.Finally,the Smoothed Intersection over Union(SIoU)loss function replaces the Complete Intersection over Union(CIoU)loss function in the original model,resulting in more accurate prediction of bounding boxes and continuous model optimization.Experimental results on the High-Resolution Remote Sensing Detection(HRRSD)dataset demonstrate that the proposed DI-YOLO model outperforms mainstream target detection algorithms in terms of mean Average Precision(mAP)for optical remote sensing image detection.Furthermore,it achieves Frames Per Second(FPS)of 138.9,meeting fast and accurate detection requirements.

Correg-Yolov3:a Method for Dense Buildings Detection in High-resolution Remote Sensing Images

The exploration of building detection plays an important role in urban planning,smart city and military.Aiming at the problem of high overlapping ratio of detection frames for dense building detection in high resolution remote sensing images,we present an effective YOLOv3 framework,corner regression-based YOLOv3(Correg-YOLOv3),to localize dense building accurately.This improved YOLOv3 algorithm establishes a vertex regression mechanism and an additional loss item about building vertex offsets relative to the center point of bounding box.By extending output dimensions,the trained model is able to output the rectangular bounding boxes and the building vertices meanwhile.Finally,we evaluate the performance of the Correg-YOLOv3 on our self-produced data set and provide a comparative analysis qualitatively and quantitatively.The experimental results achieve high performance in precision(96.45%),recall rate(95.75%),F1 score(96.10%)and average precision(98.05%),which were 2.73%,5.4%,4.1%and 4.73%higher than that of YOLOv3.Therefore,our proposed algorithm effectively tackles the problem of dense building detection in high resolution images.

Classification and Spatio-Temporal Change Detection of Land Use/Land Cover Using Remote Sensing and Geographic Information System in the Manouba Region, NE Tunisia

Land use/land cover (LULC) mapping and change detection are fundamental aspects of remote sensing data application. Therefore, selecting an appropriate classifier approach is crucial for accurate classification and change assessment. In the first part of this study, the performance of machine learning classification algorithms was compared using Landsat 9 image (2023) of the Manouba government (Tunisia). Three different classification methods were applied: Maximum Likelihood Classification (MLC), Support Vector Machine (SVM), and Random Trees (RT). The classification aimed to identify five land use classes: urban area, vegetation, bare area, water and forest. A qualitative assessment was conducted using Overall Accuracy (OA) and the Kappa coefficient (K), derived from a confusion matrix. The results of the land cover classification demonstrated a high level of accuracy. The SVM method exhibited the best performance, with an overall accuracy of 93% and a kappa accuracy of 0.9. The ML method is the second-best classifier with an overall accuracy of 92% and a kappa accuracy of 0.88. The Random Trees method yielded the lowest accuracy among the three approaches, with an overall accuracy of 91% and a kappa accuracy of 0.87. The second part of the study focused on analyzing LULC changes in the study area. Based on the classification results, the SVM method was chosen to classify the Landsat 7 image acquired in 2000. LULC changes from 2000 to 2023 were investigated using change detection comparison. The findings indicate that over the last 23 years, vegetation land and urban areas in the study area have experienced significant increases of 31.94% and 5.47%, respectively. This study contributed to a better understanding of the classification process and dynamic LULC changes in the Manouba region. It provided valuable insights for decision-makers in planning land conservation and management.

Analysis of Land Use Change and Driving Force of Bole City Based on Remote Sensing Image

[Objectives] The land use change and its influence has been the frontier and hotspot in the research of the surface process of change. The aim of this study was to provide a reasonable scientific basis for the more reasonable use of regional land resources of Bole City by study of land use change and driving force of Bole City.[Methods] Through geometric correction, image mosaic and image registration processing and classification of the remote sensing images of Bole City in 2006, 2011 and 2016, the three images of land use change in land use types （land use change range, dynamic degree and variation degree） were studied, and the natural and social economy in terms of the driving forces of land use change were analyzed.[Results] In the 2006 to 2016 period, cultivated land of Bole City had the land use dynamic growth state, and the average growth rate was 0.26%; and forest land, construction land, water, grassland and unused land showed a decreasing trend, decreased by 0.23%, 0.22%, 0.75%, 3.85% and 1.52%, respectively. In the entire study period, the change of grassland was the biggest, the changes of unused land and water were the second, and the changes of cultivated land, construction land and forest land were lesser.[Conclusions] The main driving factors that effected on land use change of the study area were climate, industrialization, urbanization, social and economic activities, adjustment of agricultural structure and population expansion.

Object Detection in Remote Sensing Images Using Picture Fuzzy Clustering and MapReduce

In image processing, one of the most important steps is image segmentation. The objects in remote sensing images often have to be detected in order toperform next steps in image processing. Remote sensing images usually havelarge size and various spatial resolutions. Thus, detecting objects in remote sensing images is very complicated. In this paper, we develop a model to detectobjects in remote sensing images based on the combination of picture fuzzy clustering and MapReduce method (denoted as MPFC). Firstly, picture fuzzy clustering is applied to segment the input images. Then, MapReduce is used to reducethe runtime with the guarantee of quality. To convert data for MapReduce processing, two new procedures are introduced, including Map_PFC and Reduce_PFC.The formal representation and details of two these procedures are presented in thispaper. The experiments on satellite image and remote sensing image datasets aregiven to evaluate proposed model. Validity indices and time consuming are usedto compare proposed model to picture fuzzy clustering model. The values ofvalidity indices show that picture fuzzy clustering integrated to MapReduce getsbetter quality of segmentation than using picture fuzzy clustering only. Moreover,on two selected image datasets, the run time of MPFC model is much less thanthat of picture fuzzy clustering.

A transformer-based Siamese network and an open optical dataset for semantic change detection of remote sensing images

Recent change detection(CD)methods focus on the extraction of deep change semantic features.However,existing methods overlook the fine-grained features and have the poor ability to capture long-range space–time information,which leads to the micro changes missing and the edges of change types smoothing.In this paper,a potential transformer-based semantic change detection(SCD)model,Pyramid-SCDFormer is proposed,which precisely recognizes the small changes and fine edges details of the changes.The SCD model selectively merges different semantic tokens in multi-head self-attention block to obtain multiscale features,which is crucial for extraction information of remote sensing images(RSIs)with multiple changes from different scales.Moreover,we create a well-annotated SCD dataset,Landsat-SCD with unprecedented time series and change types in complex scenarios.Comparing with three Convolutional Neural Network-based,one attention-based,and two transformer-based networks,experimental results demonstrate that the Pyramid-SCDFormer stably outperforms the existing state-of-the-art CD models and obtains an improvement in MIoU/F1 of 1.11/0.76%,0.57/0.50%,and 8.75/8.59%on the LEVIR-CD,WHU_CD,and Landsat-SCD dataset respectively.For change classes proportion less than 1%,the proposed model improves the MIoU by 7.17–19.53%on Landsat-SCD dataset.The recognition performance for small-scale and fine edges of change types has greatly improved.

Road Vector Map Change Monitoring Based on High Resolution Remote Sensing Images

Some studies about road vector map change detection were done in this paper. Firstly, on the basis of old road vector data, the original high resolution remote sensing image was cut into segments. Then, gray analysis and edge extraction of those segments were done so that changes of roads could be detected. Finally, according to the vector data and gray information of roads which were not changed, road templates were extracted and saved automatically. This method was performed on the World View high resolution image of certain parts in the country. The detection result shows that detection correctness is 79.56% and completeness can reach 97.72%. Moreover, the extracted road templates are essentials for the template matching method of road extraction.

Salient Object Detection from Multi-spectral Remote Sensing Images with Deep Residual Network

alient object detection aims at identifying the visually interesting object regions that are consistent with human perception. Multispectral remote sensing images provide rich radiometric information in revealing the physical properties of the observed objects, which leads to great potential to perform salient object detection for remote sensing images. Conventional salient object detection methods often employ handcrafted features to predict saliency by evaluating the pixel-wise or superpixel-wise contrast. With the recent use of deep learning framework, in particular, fully convolutional neural networks, there has been profound progress in visual saliency detection. However, this success has not been extended to multispectral remote sensing images, and existing multispectral salient object detection methods are still mainly based on handcrafted features, essentially due to the difficulties in image acquisition and labeling. In this paper, we propose a novel deep residual network based on a top-down model, which is trained in an end-to-end manner to tackle the above issues in multispectral salient object detection. Our model effectively exploits the saliency cues at different levels of the deep residual network. To overcome the limited availability of remote sensing images in training of our deep residual network, we also introduce a new spectral image reconstruction model that can generate multispectral images from RGB images. Our extensive experimental results using both multispectral and RGB salient object detection datasets demonstrate a significant performance improvement of more than 10% improvement compared with the state-of-the-art methods.

Sub-pixel change detection for urban land-cover analysis via multi-temporal remote sensing images

Conventional change detection approaches are mainly based on per-pixel processing,which ignore the sub-pixel spectral variation resulted from spectral mixture.Especially for medium-resolution remote sensing images used in urban landcover change monitoring,land use/cover components within a single pixel are usually complicated and heterogeneous due to the limitation of the spatial resolution.Thus,traditional hard detection methods based on pure pixel assumption may lead to a high level of omission and commission errors inevitably,degrading the overall accuracy of change detection.In order to address this issue and find a possible way to exploit the spectral variation in a sub-pixel level,a novel change detection scheme is designed based on the spectral mixture analysis and decision-level fusion.Nonlinear spectral mixture model is selected for spectral unmixing,and change detection is implemented in a sub-pixel level by investigating the inner-pixel subtle changes and combining multiple composition evidences.The proposed method is tested on multi-temporal Landsat Thematic Mapper and China–Brazil Earth Resources Satellite remote sensing images for the land-cover change detection over urban areas.The effectiveness of the proposed approach is confirmed in terms of several accuracy indices in contrast with two pixel-based change detection methods(i.e.change vector analysis and principal component analysis-based method).In particular,the proposed sub-pixel change detection approach not only provides the binary change information,but also obtains the characterization about change direction and intensity,which greatly extends the semantic meaning of the detected change targets.

A Summary of Change Detection Technology of Remotely-Sensed Image

This paper will describe three aspects of change detection technology of remotely-sensed images. At first, the process of change detection is presented. Then, the author makes a summary of several common change detection methods and a brief review of the advantages and disadvantages of them. At the end of this paper, the applications and difficulty of current change detection techniques are discussed.

Multi-class change detection of remote sensing images based on class rebalancing

Multi-class change detection can make various ground monitoring projects more efficient and convenient.With the development of deep learning,the multi-class change detection methods have introduced Deep Neural Network(DNN)to improve the accuracy and efficiency of traditional methods.The class imbalance in the image will affect the feature extraction effect of DNN.Existing deep learning methods rarely consider the impact of data on DNN.To solve this problem,this paper proposes a class rebalancing algorithm based on data distribution.The algorithm iteratively trains the SSL model,obtains the distribution of classes in the data,then expands the original dataset according to the distribution of classes,and finally trains the baseline SSL model using the expanded dataset.The trained semantic segmentation model is used to detect multi-class changes in two-phase images.This paper is the first time to introduce the image class balancing method in the multi-class change detection task,so a control experiment is designed to verify the effectiveness and superiority of this method for the unbalanced data.The mIoU of the class rebalancing algorithm in this paper reaches 0.4615,which indicates that the proposed method can effectively detect ground changes and accurately distinguish the types of ground changes.

Automatic Change Detection of Geo-spatial Data from Imagery

The problems and difficulty of current change detection techniques are presented. Then, according to whether image registration is done before change detection algorithms, the authors classify the change detection into two categories:the change detection after image registration and the change detection simultaneous with image registration. For the former, four topics including the change detection between new image and old image, the change detection between new image and old map, the change detection between new image/old image and old map, and the change detection between new multi-source images and old map/image are introduced. For the latter, three categories, i.e. the change detection between old DEM, DOM and new non-rectification image, the change detection between old DLG, DRG and new non-rectification image, and the 3D change detection between old 4D products and new multi-overlapped photos, are discussed.

Smart Photogrammetric and Remote Sensing Image Processing for Very High Resolution Optical Images——Examples from the CRC-AGIP Lab at UNB

This paper introduces some of the image processing techniques developed in the Canada Research Chair in Advanced Geomatics Image Processing Laboratory (CRC-AGIP Lab) and in the Department of Geodesy and Geomatics Engineering (GGE) at the University of New Brunswick (UNB), Canada. The techniques were developed by innovatively/“smartly” utilizing the characteristics of the available very high resolution optical remote sensing images to solve important problems or create new applications in photogrammetry and remote sensing. The techniques to be introduced are: automated image fusion (UNB-PanSharp), satellite image online mapping, street view technology, moving vehicle detection using single set satellite imagery, supervised image segmentation, image matching in smooth areas, and change detection using images from different viewing angles. Because of their broad application potential, some of the techniques have made a global impact, and some have demonstrated the potential for a global impact.

Land cover change detection in West Jilin using ETM + images

In order to assess the information content and accuracy ofLandsat ETM+ digital images in land cover change detection,change-detection techniques of image differencing,normalized difference vegetation index,principal components analysis and tasseled-cap transformation were applied to yield 13 images. These images were thresholded into change and no change areas. The thresholded images were then checked in terms of various accuracies. The experiment results show that kappa coefficients of the 13 images range from 48.05 ～78.09. Different images do detect different types of changes. Images associated with changes in the near-infrared-reflectance or greenness detects crop-type changes and changes between vegetative and non-vegetative features. A unique means of using only Landsat imagery without reference data for the assessment of change in arid land are presented. Images of 12th June, 2000 and 2nd June, 2002 are used to validate the means. Analyses of standard accuracy and spatial agreement are performed to compare the new images (hereafter called "change images" ) representing the change between the two dates. Spatial agreement evaluates the conformity in the classified "change pixels" and "no-change pixels" at the same location on different change images and comprehensively examines the different techniques. This method would enable authorities to monitor land degradation efficiently and accurately.

ResCD-FCN:Semantic Scene Change Detection Using Deep Neural Networks

Semantic change detection is extension of change detection task in which it is not only used to identify the changed regions but also to analyze the land area semantic(labels/categories)details before and after the timelines are analyzed.Periodical land change analysis is used for many real time applications for valuation purposes.Majority of the research works are focused on Convolutional Neural Networks(CNN)which tries to analyze changes alone.Semantic information of changes appears to be missing,there by absence of communication between the different semantic timelines and changes detected over the region happens.To overcome this limitation,a CNN network is proposed incorporating the Resnet-34 pre-trained model on Fully Convolutional Network(FCN)blocks for exploring the temporal data of satellite images in different timelines and change map between these two timelines are analyzed.Further this model achieves better results by analyzing the semantic information between the timelines and based on localized information collected from skip connections which help in generating a better change map with the categories that might have changed over a land area across timelines.Proposed model effectively examines the semantic changes such as from-to changes on land over time period.The experimental results on SECOND(Semantic Change detectiON Dataset)indicates that the proposed model yields notable improvement in performance when it is compared with the existing approaches and this also improves the semantic segmentation task on images over different timelines and the changed areas of land area across timelines.

Change Detection of Land Use and Land Cover over a Period of 20 Years in Papua New Guinea

People have an inherent tenacity to throng coastal regions in pursuit of better living conditions. As such the brisk dynamism of land use/land cover activities in a coastal region becomes obvious. The former keeps changing rapidly due to burgeoning population. A digital change detection analysis is performed with the help of Geographic Information System (GIS) on the Remote Sensing data spanning over last 20 years, complemented by in-situ data and ground truth information. This current research briefly endeavours to find out the nature of change happening in the major three coastal cities of Papua New Guinea (PNG), namely Alotau, capital of Milnebay province;Lae, capital of Morobe province and Port Moresby, capital of Papua New Guinea. Changes in land use and land cover that took place over 20 years have been recorded using Landsat 5 thematic mapper (TM) data of 1992 and Landsat 8 operational land imager (OLI) data. Land use and land cover maps of 1992, and 2013/14, and change detection matrix of 1992-2013/14 are derived. Results show an immensely sprawling urban landscape, evincing about five times growth during 1992 to 2014. At the same time “natural forests” dwindled by 444.96 hectares in Alotau, 6977.25 hectares in Lae and “mangrove” and “grass/shrub land” decreased by 127.78 and 4859.39 hectares respectively around Port Moresby. The above changes owe to ever increasing population pressure, land tenure shift, agriculture and industrial development.

RepDDNet:a fast and accurate deforestation detection model with high-resolution remote sensing image

Forest is the largest carbon reservoir and carbon absorber on earth.Thus,mapping forest cover change accurately is of great significance to achieving the global carbon neutrality goal.Accurate forest change information could be acquired by deep learning methods using high-resolution remote sensing images.However,deforestation detection based on deep learning on a large-scale region with high-resolution images required huge computational resources.Therefore,there was an urgent need for a fast and accurate deforestation detection model.In this study,we proposed an interesting but effective re-parameterization deforestation detection model,named RepDDNet.Unlike other existing models designed for deforestation detection,the main feature of RepDDNet was its decoupling feature,which means that it allowed the multi-branch structure in the training stages to be converted into a plain structure in the inference stage,thus the computation efficiency can be significantly improved in the inference stage while maintaining the accuracy unchanged.A large-scale experiment was carried out in Ankang city with 2-meter high-resolution remote sensing images(the total area of it was over 20,000 square kilometers),and the result indicated that the model computation efficiency could be improved by nearly 30%compared with the model without re-parameterization.Additionally,compared with other lightweight models,RepDDNet also displayed a trade-off between accuracy and computation efficiency.

GMTS: GNN-based multi-scale transformer siamese network for remote sensing building change detection

With the remarkable success of change detection(CD)in remote sensing images in the context of deep learning,many convolutional neural network(CNN)based methods have been proposed.In the current research,to obtain a better context modeling method for remote sensing images and to capture more spatiotemporal characteristics,several attention-based methods and transformer(TR)-based methods have been proposed.Recent research has also continued to innovate on TR-based methods,and many new methods have been proposed.Most of them require a huge number of calculation to achieve good results.Therefore,using the TR-based mehtod while maintaining the overhead low is a problem to be solved.Here,we propose a GNN-based multi-scale transformer siamese network for remote sensing image change detection(GMTS)that maintains a low network overhead while effectively modeling context in the spatiotemporal domain.We also design a novel hybrid backbone to extract features.Compared with the current CNN backbone,our backbone network has a lower overhead and achieves better results.Further,we use high/low frequency(HiLo)attention to extract more detailed local features and the multi-scale pooling pyramid transformer(MPPT)module to focus on more global features respectively.Finally,we leverage the context modeling capabilities of TR in the spatiotemporal domain to optimize the extracted features.We have a relatively low number of parameters compared to that required by current TR-based methods and achieve a good effect improvement,which provides a good balance between efficiency and performance.

Multi-temporal urban semantic understanding based on GF-2 remote sensing imagery:from tri-temporal datasets to multi-task mapping

High resolution satellite images are becoming increasingly available for urban multi-temporal semantic understanding.However,few datasets can be used for land-use/land-cover(LULC)classification,binary change detection(BCD)and semantic change detection(SCD)simultaneously because classification datasets always have one time phase and BCD datasets focus only on the changed location,ignoring the changed classes.Public SCD datasets are rare but much needed.To solve the above problems,a tri-temporal SCD dataset made up of Gaofen-2(GF-2)remote sensing imagery(with 11 LULC classes and 60 change directions)was built in this study,namely,the Wuhan Urban Semantic Understanding(WUSU)dataset.Popular deep learning based methods for LULC classification,BCD and SCD are tested to verify the reliability of WUSU.A Siamese-based multi-task joint framework with a multi-task joint loss(MJ loss)named ChangeMJ is proposed to restore the object boundaries and obtains the best results in LULC classification,BCD and SCD,compared to the state-of-the-art(SOTA)methods.Finally,a large spatial-scale mapping for Wuhan central urban area is carried out to verify that the WUsU dataset and the ChangeMJ framework have good application values.