ZY-3 Images Research Articles

A multi-scale weakly supervised learning method with adaptive online noise correction for high-resolution change detection of built-up areas

Accurate change detection of built-up areas (BAs) fosters a comprehensive understanding of urban development. The post-classification comparison (PCC) is a widely-used change detection method by classification and temporal comparison. For classification, image-level labeling is an efficient alternative to pixel-level one for pixel-wise weakly supervised segmentation, which frequently applies pixel-level pseudo labels generated from class activation map (CAM) to train semantic segmentation networks. CAM can be obtained from classification networks trained with image-level labels and can indicate the spatial location of objects. The existing studies are subject to the following issues: 1) They only rely on the single-scale and low-resolution CAM, but ignore the multi-scale property of BAs; 2) Pixel-level pseudo labels usually contain noises (e.g., omissions and false alarms); 3) The temporal correlation between multi-temporal images is less considered in PCC. To address these limitations, this paper proposed a multi-scale weakly supervised learning method, which utilized a large number of single-temporal high-resolution images and image-level labels to detect BA changes. This method consisted of three modules: 1) multi-scale CAM for BA pseudo label generation; 2) adaptive online noise correction for BA detection; and 3) generation of reliable pseudo labels for BA change detection. Based on ZY-3 images (2.5 m), we constructed the first multi-view datasets for both BA detection and change detection. Each ZY-3 image includes a multi-spectral image with red, green, blue, and near-infrared bands and a multi-view image with nadir-, forward-, and backward-views. The BA detection dataset contained 86,166 image-level samples (256 × 256 pixels for each sample), covering 48 major cities in China, while the BA change detection dataset consisted of ZY-3 bi-temporal images at rapidly urbanizing areas (i.e., Beijing and Shanghai). Experiments showed that the proposed method can detect BA changes and suppress pseudo changes effectively, yielding 88.2% F1-score in BA detection and 79.3% for Shanghai and 78.5% for Beijing in change detection. Further analysis demonstrated the proposed method to be advantageous in the following two fronts: 1) the image-level weak labels can achieve pixel-wise BA change detection at low cost; and 2) the multi-scale CAM and temporal correlation are effective in the scenarios with limited labels. Datasets and codes will be accessed at https://github.com/lauraset/MSWS.

A full-level fused cross-task transfer learning method for building change detection using noise-robust pretrained networks on crowdsourced labels

Accurate building change detection is crucial for understanding urban development. Although fully supervised deep learning-based methods for building change detection have made progress, they tend to fuse temporal information only at a single level (e.g., input, feature, or decision levels) to mitigate the data distribution differences between time-series images, which is highly prone to introduce a large number of pseudo changes. Moreover, these methods rely on a large number of high-quality pixel-level change labels with high acquisition costs. In contrast, available crowdsourced building data are abundant but are less considered for change detection. For example, OpenStreetMap (OSM), Google Map, and Gaode Map provide lots of available building labels, yet they usually contain noise such as false alarms, omissions, and mismatches, limiting their wide application. In addition, when the building extraction task is transferred to the building change detection task, the temporal and regional differences between different images may lead to undesired pseudo changes. Given these issues, we propose a full-level fused cross-task transfer learning method for building change detection using only crowdsourced building labels and high-resolution satellite imagery. The method consists of three steps: 1) noise-robust building extraction network pretraining; 2) uncertainty-aware pseudo label generation; and 3) full-level fused building change detection. We created building extraction and building change detection datasets. The former (building extraction dataset) contains 30 scenes of ZY-3 images covering 27 major cities in China and crowdsourced building labels from Gaode Map for training, while the latter (building change dataset) contains bi-temporal ZY-3 images in Shanghai and Beijing for testing. The results show that the proposed method can identify changed buildings more effectively and better balance false alarms and omissions, compared to the existing state-of-the-art methods. Further analysis indicates that the inclusion of samples from multiple cities with various spatial heterogeneities is helpful to improve the performance. The experiments show that it is promising to apply the proposed method to situations where true labels are completely lacking or limited, thus alleviating the issue of high label acquisition cost. The source code will be available at https://github.com/lauraset/FFCTL.

Identifying Coastal Wetlands Changes Using a High-Resolution Optical Images Feature Hierarchical Selection Method

Coastal wetlands are dynamic and fragile ecosystems where complex changes have taken place. As they are affected by environmental changes and human activities, it is of great practical significance to monitor coastal wetlands changes regularly. High-resolution optical data can observe changes in coastal wetlands, however, the impact of different optical features on the identification of changes in coastal wetlands is not clear. Simultaneously, the combination of many features could cause the “dimension disaster” problem. In addition, only small amounts of training samples are accessible at pre- or post-changed time. In order to solve the above problems, the feature hierarchical selection method is proposed, taking into account the jumping degree of different image features. The influence of different optical features on wetland classification was analyzed. In addition, a training samples transfer learning strategy was designed for wetland classification, and the classification result at pre- and post-changed times were compared to identify the “from-to” coastal wetlands changes. The southeastern coastal wetlands located in Jiangsu Province were used as a study area, and ZY-3 images in 2013 and 2018 were used to verify the proposed methods. The results show that the feature hierarchical selection method can provide a quantitative reference for optimal subset feature selection. A training samples transfer learning strategy was used to classify post-changed optical data, the overall accuracy of transferred training samples was 91.16%, and it ensures the accuracy requirements for change identification. In the study area, the salt marsh increased mainly from the sea area, because salt marshes expand rapidly throughout coastal areas, and aquaculture ponds increased from the sea area and salt marshes, because of the considerable economic benefits of the aquacultural industry.

Remote sensing evaluation of ecological restoration engineering effect: A case study of the Yongding River Watershed, China

Ecological restoration has been risen to a national strategic project for China and become an significant way to realize ecological civilization construction. As an important way to measure the effectiveness of ecological restoration engineering, the corresponding evaluation provides a reliable basis for ecological restoration decision-making and ecosystem management. Based on remote sensing technology with the advantages of short revisit time, large monitoring range, high spatial resolution, this paper proposes a remote sensing evaluation method of watershed ecological restoration effect, so as to monitor the implementation effect of ecological restoration engineering in the watershed and technically check whether the expected goals set before the restoration are achieved. The comprehensive ecological restoration engineering of Yongding River has been implemented since 2016, and it was the first focus of the coordinated development of Beijing-Tianjin-Hebei region to make advances in ecology. In this paper, year 2015 and 2020 were selected as the before and the after the ecological restoration engineering of Yongding River. The research object was the riparian zone from Luopoling Reservoir to Daning Reservoir in Yongding River Watershed. After obtaining and processing remote sensing images, ZY-3 images were used to obtain land cover data by means of human-computer interactive interpretation, and Landsat images were used to retrieve NDVI data. Based on these data, the changes of the eight indicators of water resources, water ecology and ecological function in the study area were calculated to evaluate the ecological restoration engineering effect. The results showed that the study area has basically realized the maintenance of ecological water quantity, formed the ecological corridor of Yongding River, and effectively enhanced the ecological function and improved the water environment. By verifying the feasibility of remote sensing evaluation method, this paper could provide relevant decision makers and planners with an effective evaluation method for the ecological restoration engineering effect based on remote sensing data and technology.

Estimating Particulate Matter Emission from Dust Sources Using ZY-3 Data and GIS Technology—A Case Study in Zhengzhou City, China

With the rapid development of the social economy in China, numerous Chinese cities are facing high levels of particulate matter (PM) pollution problems. In this study, high-resolution ZY-3 images and GIS techniques were used to establish the emission inventory of total suspended particle (TSP), particulate matter 10 (PM10) and particulate matter 2.5 (PM2.5) from fugitive dust sources in May 2016, and a spatial grid of 3 km × 3 km resolution was established to demonstrate the spatial distribution of PM emission. Results showed that the total emissions of TSP, PM10 and PM2.5 in Zhengzhou city were 237.5 kt·a−1, 103.7 kt·a−1 and 22.4 kt·a−1, respectively. Construction dust source was the main fugitive dust emission source in Zhengzhou city—the TSP, PM10 and PM2.5 emission of which account for 76.42%, 89.68% and 88.39%, respectively, of the total emission, followed by road dust source and soil dust source. PM emission was higher in Zhongyuan, Huiji, Jinshui and Zhengdong New District, while Zhongmou, Xingyang, Dengfeng and other remote areas had low PM emissions. Compared to other Chinese cities or regions, the PM emission from the construction dust source was at a high level in Zhengzhou city, while the PM emissions from the soil dust source and road dust source were at moderate levels.

Nonlinear Systematic Distortions Compensation in Satellite Images Based on an Equivalent Geometric Sensor Model Recovered From RPCs

The rational polynomial function is widely accepted as the preferred sensor model for high-resolution satellite imagery (HRSI). However, satellite images and their associated rational polynomial coefficients (RPCs) often suffer from nonlinear systematic errors which were caused by attitude oscillation, sensor deformation, and many other imperfect calibration errors, thus affecting the geo-referencing accuracy. Instead of modeling the biases by polynomials in the image space or refining RPCs directly, this study proposes an approach of going back to the physical model and correcting the local distortions in a self-calibration block adjustment. The algorithm of an equivalent geometric sensor model (EGSM) recovery from RPCs is described in detail. As an equivalent form of the physical sensor models, EGSM reflects the complete viewing geometry of push-broom HRSI. The interior and exterior orientation parameters of EGSM can be stably recovered from RPCs without using any metadata. An approach of RPCs refinement by self-calibration block adjustment based on EGSM is introduced. This approach can effectively compensate for the nonlinear systematic errors caused by platforms and sensors similar to the approach of a rigorous sensor model. The performance of EGSM-based block adjustment is compared with the RFM-based bias compensation method. Experiments using ZY-3 images show the EGSM-based approach can effectively eliminate the nonlinear distortions in satellite images caused by sensor deformation and attitude vibrations. Furthermore, experiments using images from various satellites show that the original RFM can be well fitted with the EGSM and the residuals are smaller than 0.1 pixels for all test images.

Hierarchical Geographic Object-Based Vegetation Type Extraction Based on Multi-Source Remote Sensing Data

Providing vegetation type information with accurate surface distribution is one of the important tasks of remote sensing of the ecological environment. Many studies have explored ecosystem structure information at specific spatial scales based on specific remote sensing data, but it is still rare to extract vegetation information at various landscape levels from a variety of remote sensing data. Based on Gaofen-1 satellite (GF-1) Wide-Field-View (WFV) data (16 m), Ziyuan-3 satellite (ZY-3) and airborne LiDAR data, this study comparatively analyzed the four levels of vegetation information by using the geographic object-based image analysis method (GEOBIA) on the typical natural secondary forest in Northeast China. The four levels of vegetation information include vegetation/non-vegetation (L1), vegetation type (L2), forest type (L3) and canopy and canopy gap (L4). The results showed that vegetation height and density provided by airborne LiDAR data could extract vegetation features and categories more effectively than the spectral information provided by GF-1 and ZY-3 images. Only 0.5 m LiDAR data can extract four levels of vegetation information (L1–L4); and from L1 to L4, the total accuracy of the classification decreased orderly 98%, 93%, 80% and 69%. Comparing with 2.1 m ZY-3, the total classification accuracy of L1, L2 and L3 extracted by 2.1 m LiDAR data increased by 3%, 17% and 43%, respectively. At the vegetation/non-vegetation level, the spatial resolution of data plays a leading role, and the data types used at the vegetation type and forest type level become the main influencing factors. This study will provide reference for data selection and mapping strategies for hierarchical multi-scale vegetation type extraction.

COMPREHENSIVE ANALYSIS OF URBAN THERMAL ENVIRONMENT – A CASE STUDY OF QINGDAO，CHINA

Abstract. The thermal environment is an important part of spatial planning of national land. It can be inverted by thermal infrared remote sensing image data. Qingdao is an important coastal city in China which is effected by the urban heat island. The driving factors of urban thermal environment in Qingdao are analysed by using Landsat 8 OLI/TIRS images, DMSP/OLS data, ZY-3 images, statistical data et al. Firstly, the urban heat island information in different periods are retrieved by Landsat8 TIRS by the surface temperature. On this basis, the coupling analysis of the characteristics of vegetation cover, water, land use type, night light, urban impervious surface, transportation, change density and other factors that related to spatial planning of national land and the change of thermal environment is completed from two dimensions of time and space. At the same time, according to the spatial distribution of the heat island evolution, with the aim of improving the thermal environment and optimizing the urban spatial layout of Qingdao, the paper puts forward reasonable suggestions and improvement strategies, and provides basic reference data for the development of urban heat island governance and land development and utilization.

STUDY OF LARGE-SCALE BLOCK ADJUSTMENT WITH LASER ALTIMETRY DATA GLAS OF ZY3 IMAGES OVER SHANDONG

Abstract. In the global mapping work, there will be no control points or uneven distribution of control points. The traditional method of single scene image correction based on ground control points is not suitable for modern surveying and mapping, which has no available control points in a large range. New control data is urgently needed to make up for the influence of traditional control points on surveying and mapping. In this paper, aiming at improving the elevation accuracy of ZY-3 uncontrolled point surveying, a filtering principle based on the analysis based on waveform analysis and the stereo image pair based on tie points is proposed for the screening of GLAS laser elevation points. A new method using high-reliability laser elevation control points as the height constraint and the regional network adjustment of ZY-3 images without control points is studied. Through the experiment in Shandong Province, it is proved that the elevation accuracy of the adjustment can be improved obviously by introducing the selected laser points. The mean error of adjustment elevation is increased from −2.297 to 0.216m and the mean square error of elevation is increased from 3.193m to 2.007m, significantly improving the elevation positioning accuracy of adjustment.

Extraction of raft cultivation of remotely sensed high-spatial-resolution images based on LBV and wavelet transforms

A method to extract information on raft cultivation based on the LBV (The general radiance level L, visible-infrared radiation balance B, and band radiance variation vector V) transform and wavelet multi-scale decomposition was proposed. This paper selects the spectral information of typical objects and uses regression analysis to solve for the coefficients of regression. The LBV transformation of remotely sensed high-spatial-resolution imagery was derived according to the spatial characteristics of the object space, color space, and LBV variable space to obtain the B band that can be used to identify maritime targets. Multi-scale wavelet decomposition was then used to calculate the eigenvalues of the components of each sub-image by using different windows to analyze the texture features of the maritime objects, and to explore the rules of texture extraction of raft cultivation from high-spatial-resolution images. Experiments on remotely sensed ZY-3 images from Liandao Island show the following: (1) compared with the original image, the LBV-transformed image was clearer and finer details were visible in it. They helped identify and extract ground objects. (2) The eigenvalues extracted by wavelet decomposition were beneficial for the segmentation of the image. As the dimensionality of the eigenvalue increased, the accuracy of raft cultivation extraction showed an upward trend. (3) The proposed method is applicable to texture feature extraction owing to its clear rules and directionality.

Long-short-term-memory-based crop classification using high-resolution optical images and multi-temporal SAR data

Farmland parcel-based crop classification using satellite data plays an important role in precision agriculture. In this study, a deep-learning-based time-series analysis method employing optical images and synthetic aperture radar (SAR) data is presented for crop classification for cloudy and rainy regions. Central to this method is the spatial-temporal incorporation of high-resolution optical images and multi-temporal SAR data and deep-learning-based time-series analysis. First, a precise farmland parcel map is delineated from high-resolution optical images. Second, pre-processed SAR intensity images are overlaid onto the parcel map to construct time series of crop growth for each parcel. Third, a deep-learning-based (using the long short-term memory, LSTM, network) classifier is employed to learn time-series features of crops and to classify parcels to produce a final classification map. The method was applied to two datasets of high-resolution ZY-3 images and multi-temporal Sentinel-1A SAR data to classify crop types in Hunan and Guizhou of China. The classification results, with an 5.0% improvement in overall accuracy compared to those of traditional methods, illustrate the effectiveness of the proposed framework for parcel-based crop classification for southern China. A further analysis of the relationship between crop calendars and change patterns of time-series intensity indicates that the LSTM model could learn and extract useful features for time-series crop classification.

MULTI-SOURCE SATELLITE OBSERVATIONS REVEAL EVOLUTION PATTERN OF RIFTS IN THE FILCHNER-RONNE ICE SHELF, ANTARCTICA

Abstract. This paper presents a systematic fracturing study of the Filchner-Ronne Ice Shelf (FRIS) in Antarctica based on multisource data dating back to 1973. Two fracture maps of FRIS were extracted from Landsat-8 OLI images of 2014–2015 and MODIS mosaic of Antarctica (MOA) 2008–2009. The evolution pattern and temporal change of two crucial rifts, Rifts T1 and T2 are analysed for their similarities with the Grand Chasm, which directly induced the previous major calving event of Filchner Ice Shelf (FIS) in 1986. The depth temporal change of Rift T2 was also analysed based on ICESat altimetry data and DEMs reconstructed from ZY-3 and WV-2 stereo images. 3D parameters were also extracted and analysed from DEMs, including rift depth, walls, mélange surface roughness, and mélange thickness. In total, 582 new fractures were extracted from 2004 to 2009 and another 752 from 2008 to 2015. Rifts T1 and T2 showed high activeness, which experienced a rapid growth of 82% and reached ∼50 km in length. Based on all the observations of fracture and rift activity, some regions of ice shelf front in RIS and FIS show some important characteristics that can be related back to the previous calving events.

THE APPLICATION OF CHINESE HIGH-SPATIAL-RESOLUTION REMOTE SENSING SATELLITE IMAGE IN LAND LAW ENFORCEMENT INFORMATION EXTRACTION

Abstract. Chinese high -resolution (HR) remote sensing satellites have made huge leap in the past decade. Commercial satellite datasets, such as GF-1, GF-2 and ZY-3 images, the panchromatic images (PAN) resolution of them are 2 m, 1 m and 2.1 m and the multispectral images (MS) resolution are 8 m, 4 m, 5.8 m respectively have been emerged in recent years. Chinese HR satellite imagery has been free downloaded for public welfare purposes using. Local government began to employ more professional technician to improve traditional land management technology. This paper focused on analysing the actual requirements of the applications in government land law enforcement in Guangxi Autonomous Region. 66 counties in Guangxi Autonomous Region were selected for illegal land utilization spot extraction with fusion Chinese HR images. The procedure contains: A. Defines illegal land utilization spot type. B. Data collection, GF-1, GF-2, and ZY-3 datasets were acquired in the first half year of 2016 and other auxiliary data were collected in 2015. C. Batch process, HR images were collected for batch preprocessing through ENVI/IDL tool. D. Illegal land utilization spot extraction by visual interpretation. E. Obtaining attribute data with ArcGIS Geoprocessor (GP) model. F. Thematic mapping and surveying. Through analysing 42 counties results, law enforcement officials found 1092 illegal land using spots and 16 suspicious illegal mining spots. The results show that Chinese HR satellite images have great potential for feature information extraction and the processing procedure appears robust.

SUGARCANE CROP EXTRACTION USING OBJECT-ORIENTED METHOD FROM ZY-3 HIGH RESOLUTION SATELLITE TLC IMAGE

Abstract. Sugarcane is one of the most important crops in Guangxi, China. As the development of satellite remote sensing technology, more remotely sensed images can be used for monitoring sugarcane crop. With the help of Three Line Camera (TLC) images, wide coverage and stereoscopic mapping ability, Chinese ZY-3 high resolution stereoscopic mapping satellite is useful in attaining more information for sugarcane crop monitoring, such as spectral, shape, texture difference between forward, nadir and backward images. Digital surface model (DSM) derived from ZY-3 TLC images are also able to provide height information for sugarcane crop. In this study, we make attempt to extract sugarcane crop from ZY-3 images, which are acquired in harvest period. Ortho-rectified TLC images, fused image, DSM are processed for our extraction. Then Object-oriented method is used in image segmentation, example collection, and feature extraction. The results of our study show that with the help of ZY-3 TLC image, the information of sugarcane crop in harvest time can be automatic extracted, with an overall accuracy of about 85.3 %.

BUILT-UP AREA DETECTION FROM HIGH-RESOLUTION SATELLITE IMAGES USING MULTI-SCALE WAVELET TRANSFORM AND LOCAL SPATIAL STATISTICS

Abstract. Recently, built-up area detection from high-resolution satellite images (HRSI) has attracted increasing attention because HRSI can provide more detailed object information. In this paper, multi-resolution wavelet transform and local spatial autocorrelation statistic are introduced to model the spatial patterns of built-up areas. First, the input image is decomposed into high- and low-frequency subbands by wavelet transform at three levels. Then the high-frequency detail information in three directions (horizontal, vertical and diagonal) are extracted followed by a maximization operation to integrate the information in all directions. Afterward, a cross-scale operation is implemented to fuse different levels of information. Finally, local spatial autocorrelation statistic is introduced to enhance the saliency of built-up features and an adaptive threshold algorithm is used to achieve the detection of built-up areas. Experiments are conducted on ZY-3 and Quickbird panchromatic satellite images, and the results show that the proposed method is very effective for built-up area detection.

Matching of Remote Sensing Images with Complex Background Variations via Siamese Convolutional Neural Network

Feature-based matching methods have been widely used in remote sensing image matching given their capability to achieve excellent performance despite image geometric and radiometric distortions. However, most of the feature-based methods are unreliable for complex background variations, because the gradient or other image grayscale information used to construct the feature descriptor is sensitive to image background variations. Recently, deep learning-based methods have been proven suitable for high-level feature representation and comparison in image matching. Inspired by the progresses made in deep learning, a new technical framework for remote sensing image matching based on the Siamese convolutional neural network is presented in this paper. First, a Siamese-type network architecture is designed to simultaneously learn the features and the corresponding similarity metric from labeled training examples of matching and non-matching true-color patch pairs. In the proposed network, two streams of convolutional and pooling layers sharing identical weights are arranged without the manually designed features. The number of convolutional layers is determined based on the factors that affect image matching. The sigmoid function is employed to compute the matching and non-matching probabilities in the output layer. Second, a gridding sub-pixel Harris algorithm is used to obtain the accurate localization of candidate matches. Third, a Gaussian pyramid coupling quadtree is adopted to gradually narrow down the searching space of the candidate matches, and multiscale patches are compared synchronously. Subsequently, a similarity measure based on the output of the sigmoid is adopted to find the initial matches. Finally, the random sample consensus algorithm and the whole-to-local quadratic polynomial constraints are used to remove false matches. In the experiments, different types of satellite datasets, such as ZY3, GF1, IKONOS, and Google Earth images, with complex background variations are used to evaluate the performance of the proposed method. The experimental results demonstrate that the proposed method, which can significantly improve the matching performance of multi-temporal remote sensing images with complex background variations, is better than the state-of-the-art matching methods. In our experiments, the proposed method obtained a large number of evenly distributed matches (at least 10 times more than other methods) and achieved a high accuracy (less than 1 pixel in terms of root mean square error).

Matching Multi-Source Optical Satellite Imagery Exploiting a Multi-Stage Approach

Geometric distortions and intensity differences always exist in multi-source optical satellite imagery, seriously reducing the similarity between images, making it difficult to obtain adequate, accurate, stable, and well-distributed matches for image registration. With the goal of solving these problems, an effective image matching method is presented in this study for multi-source optical satellite imagery. The proposed method includes three steps: feature extraction, initial matching, and matching propagation. Firstly, a uniform robust scale invariant feature transform (UR-SIFT) detector was used to extract adequate and well-distributed feature points. Secondly, initial matching was conducted based on the Euclidean distance to obtain a few correct matches and the initial projective transformation between the image pair. Finally, two matching strategies were used to propagate matches and produce more reliable matching results. By using the geometric relationship between the image pair, geometric correspondence matching found more matches than the initial UR-SIFT feature points. Further probability relaxation matching propagated some new matches around the initial UR-SIFT feature points. Comprehensive experiments on Chinese ZY3 and GaoFen (GF) satellite images revealed that the proposed algorithm performs well in terms of the number of correct matches, correct matching rate, spatial distribution, and matching accuracy, compared to the standard UR-SIFT and triangulation-based propagation method.

Changes in ice volume of the Ningchan No.1 Glacier, China, from 1972 to 2014, as derived from in situ measurements

ABSTRACTGlobal climate change is causing widespread glacier retreat, with many small glaciers disappearing from the world's mountain ranges. We obtained the annual mass balance of a small glacier (the Ningchan No.1 Glacier) located on the northeastern Tibetan Plateau, from the years 2010 to 2015 using glaciological and geodetic methods. We also measured the glacier's thickness in 2014 using ground-penetrating radar. Employing topographical maps and ZY-3 images, we obtained Digital Elevation Models for 1972 and 2014. Our results showed that the mean annual mass balance from 2010 to 2015 was ~−0.9 ± 0.5 m w.e. The mean equilibrium line altitude was ~4680 m in the period 2010–15, which exceeds the maximum elevation of the glacier. The glacier has lost area and mass across its elevation range. The mean ice thickness was 24.0 ± 2.5 m in 2014. From 1972 to 2014, the glacier's area shrank from 0.77 ± 0.05 to 0.39 ± 0.04 km2, and the ice volume decreased by (14.96 ± 0.97) × 106 m3, equivalent to (12.72 ± 0.82) × 106 t w.e. over the same period.

Large-scale block adjustment without use of ground control points based on the compensation of geometric calibration for ZY-3 images

The potential of large-scale block adjustment (BA) without ground control points (GCPs) has long been a concern among photogrammetric researchers, which is of effective guiding significance for global mapping. However, significant problems with the accuracy and efficiency of this method remain to be solved. In this study, we analyzed the effects of geometric errors on BA, and then developed a step-wise BA method to conduct integrated processing of large-scale ZY-3 satellite images without GCPs. We first pre-processed the BA data, by adopting a geometric calibration (GC) method based on the viewing-angle model to compensate for systematic errors, such that the BA input images were of good initial geometric quality. The second step was integrated BA without GCPs, in which a series of technical methods were used to solve bottleneck problems and ensure accuracy and efficiency. The BA model, based on virtual control points (VCPs), was constructed to address the rank deficiency problem caused by lack of absolute constraints. We then developed a parallel matching strategy to improve the efficiency of tie points (TPs) matching, and adopted a three-array data structure based on sparsity to relieve the storage and calculation burden of the high-order modified equation. Finally, we used the conjugate gradient method to improve the speed of solving the high-order equations. To evaluate the feasibility of the presented large-scale BA method, we conducted three experiments on real data collected by the ZY-3 satellite. The experimental results indicate that the presented method can effectively improve the geometric accuracies of ZY-3 satellite images. This study demonstrates the feasibility of large-scale mapping without GCPs.

Phenological metrics-based crop classification using HJ-1 CCD images and Landsat 8 imagery

ABSTRACTCrop type data are an important piece of information for many applications in agriculture. Extracting crop type using remote sensing is not easy because multiple crops are usually planted into small parcels with limited availability of satellite images due to weather conditions. In this research, we aim at producing crop maps for areas with abundant rainfall and small-sized parcels by making full use of Landsat 8 and HJ-1 charge-coupled device (CCD) data. We masked out non-vegetation areas by using Landsat 8 images and then extracted a crop map from a long-term time-series of HJ-1 CCD satellite images acquired at 30-m spatial resolution and two-day temporal resolution. To increase accuracy, four key phenological metrics of crops were extracted from time-series Normalized Difference Vegetation Index curves plotted from the HJ-1 CCD images. These phenological metrics were used to further identify each of the crop types with less, but easier to access, ancillary field survey data. We used crop area data from the Jingzhou statistical yearbook and 5.8-m spatial resolution ZY-3 satellite images to perform an accuracy assessment. The results show that our classification accuracy was 92% when compared with the highly accurate but limited ZY-3 images and matched up to 80% to the statistical crop areas.