Journals >Spacecraft Recovery & Remote Sensing
Contents
2024
Volume: 45 Issue 3
14 Article(s)
Export citation format
[in Chinese]
Analysis on the Development of Intelligent Spacecrafts
Yaoxin ZHANG, Xiaopei PAN, and Jingbo ZHANG
Artificial intelligence (AI) technology has been gradually developed and matured since its birth in 1950s. It has been widely used in different fields of society, and the major space agencies have gradually applied it into the space field. With the development and application of the intelligent technology, spacecrafts Artificial intelligence (AI) technology has been gradually developed and matured since its birth in 1950s. It has been widely used in different fields of society, and the major space agencies have gradually applied it into the space field. With the development and application of the intelligent technology, spacecrafts are gradually equipped with almost human-like behavior characteristics such as acquirement, recognition, planning, decision, control, execution, management and operation. Some spacecrafts even have group cooperation and other social characteristics, which significantly enhance the spacecraft abilities to cope with complex environment. In the paper, the connotation of intelligent spacecraft was discussed; the development course, the development characteristics and main application cases of the intelligent spacecraft were analyzed; and the development proposals for our intelligent spacecraft were put forward..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 1 (2024)
Advances in Research on Large-Aperture Optical Synthetic Aperture Systems
Jiafu ZHANG, Xiaoyong WANG, Mengxu LI, Xin TAO, Ling LI, and Renyuan WANG
Optical synthetic aperture technology is an effective way to increase the effective aperture of telescopes, enhance their resolutions, and improve their light-gathering capabilities. However, the technical difficulty of large-aperture optical synthetic aperture systems grows exponentially with the increase in sub-apertOptical synthetic aperture technology is an effective way to increase the effective aperture of telescopes, enhance their resolutions, and improve their light-gathering capabilities. However, the technical difficulty of large-aperture optical synthetic aperture systems grows exponentially with the increase in sub-aperture discreteness and the release of degrees of freedom. This paper compares the basic principles and structural forms of optical synthetic aperture systems, reviews the development history of large-aperture optical synthetic aperture systems represented by segmented, sparse aperture, and distributed configurations, and analyzes the technical challenges faced in key areas such as active optics, formation and control, and path planning. Furthermore, the development direction is proposed for the next generation of large-aperture optical synthetic aperture systems..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 7 (2024)
Design and Test for a Cryogenic Beamsplitter Support Structure with Large Aspect Ratio
Weiming TONG, Yue MA, Junru SONG, Libing JIN, and Minlong LIAN
Mounting large radius-to-thickness ratio beamsplitters for cryogenic space application requires a careful trade-off in stiffness and flexibility design. Both the cryogenic wavefront error and the survivability in launch vibration environments should be satisfied simultaneously. A new support structure is designed in thMounting large radius-to-thickness ratio beamsplitters for cryogenic space application requires a careful trade-off in stiffness and flexibility design. Both the cryogenic wavefront error and the survivability in launch vibration environments should be satisfied simultaneously. A new support structure is designed in this paper. In the design, the difference of thermal expansion coefficient of different structural materials is used to achieve athermal support. At the same time, multistage radial flexible structures are designed at the connection positions of different materials. The support structure achieves both high stiffness and low cryogenic wavefront error. The simulation results show that the first mode frequency is 191 Hz. The wavefront error changes 0.032λ from 293 K to 133 K based on FEM simulation results. After the beamsplitter is assembled, the cryogenic wavefront error and vibration tests are carried out. The wavefront error changes 0.028λ from 293 K to 130 K. The maximum acceleration response reaches 16gn at the installation position of the beamsplitter assembly. There is no change for the beamsplitter assembly before and after the vibration test. The stiffness, cryogenic wavefront error and environmental adaptability of the support structure of the cryogenic larger radius-to-thickness beamsplitter in this paper meet the application requirements of space cameras..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 19 (2024)
[in Chinese]
Infrared Small Target Detection Based on Four-Dimensional Spatial-Temporal Tensor
Yuan LUO, Xiaorun LI, Shuhan CHEN, and Chaoqun XIA
The existing infrared small target detection technologies have some shortcomings in terms of target detection capability, background suppression capability, and real-time performance, which fails to meet practical needs. Tensor analysis techniques have been widely used for infrared small target detection and have increThe existing infrared small target detection technologies have some shortcomings in terms of target detection capability, background suppression capability, and real-time performance, which fails to meet practical needs. Tensor analysis techniques have been widely used for infrared small target detection and have increasingly demonstrated superiority. However, three are three key issues, including suitable tensor structures, comprehensive tensor decomposition frameworks, and satisfactory real-time performance. Consequently, this paper proposes an infrared small target detection method based on four dimensional temporal-spatial tensor decomposition and block term decomposition-based norm (BTDN-4DST). Specifically, a four-dimensional sphered temporal-spatial image-patch tensor is firstly constructed to establish the data basis for tensor decomposition. Subsequently, a norm based on block term decomposition is defined to fully exploit the spatial-temporal characteristics of the background for accurate background estimation. Finally, an effective solution framework based on Alternating Direction Method of Multipliers is designed for solving the detection model. To validate the performance of the BTDN-4DST, six state-of-the-art infrared small target detection methods are selected as comparative algorithms, and extensive experiments and analyses are conducted on five real infrared image sequence datasets. BTDN-4DST can rapidly enhance the saliency of weak small targets and greatly suppress background and noise components, which proves that the proposed method not only excels in background suppressibility and target detectability but also exhibits satisfactory real-time detection performance, meeting practical application requirements..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 28 (2024)
Application Test of GF-3 Satellite Image in Land Surface Change Detection
Wenfei TAI, Xuhui CHEN, Xinsheng ZHANG, Mingyong CAI, Zhihua REN, Lixia WANG, and Xuewei SHI
Synthetic aperture radar (SAR) image is not widely used in the field of change detection due to coherent spot noise, imaging geometric distortion, and the complexity of background information. This paper summarizes the current commonly used SAR image-based change detection methods and technical processes, builds GaofenSynthetic aperture radar (SAR) image is not widely used in the field of change detection due to coherent spot noise, imaging geometric distortion, and the complexity of background information. This paper summarizes the current commonly used SAR image-based change detection methods and technical processes, builds Gaofen-3 SAR image difference interferometry and polarization data change detection technology methods and processes, and chooses the new well open coal mine area in Inner Mongolia, successfully verifying the technology feasibility and reliability of the application of GF-3 images to change detection . The main conclusions are as follows: 1) the deformation map of new coal mine is successfully obtained based on GF-3 satellite differential image differential interferometry, and the large deformation area basically coincides with the optical image interpretation results, proving that GF-3 satellite has differential interferometric capability; 2) the new coal mine accident location based on GF-3 image change detection is accurate, and the extracted change area is 0.128 km2, which is basically consistent with the publicly released monitoring results (0.1 km2), proving that GF-3 image can be applied to surface change detection..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 41 (2024)
Application Analysis of Object-Level (OL) Spatio-Temporal Fusion Model in NDVI and LST —— Taking Dali Area as an Example
Yongle GAO, Jinsheng CHANG, Yongchong YANG, and Tao WANG
In the past decades, spatio-temporal fusion technology has provided an economical and efficient method to realize long-time series observation, but this method has a weak ability to retain structural information and low computational efficiency. In this study, the difference of fusion effect and fusion efficiency betweIn the past decades, spatio-temporal fusion technology has provided an economical and efficient method to realize long-time series observation, but this method has a weak ability to retain structural information and low computational efficiency. In this study, the difference of fusion effect and fusion efficiency between mainstream spatiotemporal fusion methods (including STARFM, ESTARFM, Fit-FC, FSDAF) and Object-level spatiotemporal fusion methods in normalized vegetation index (NDVI) and land surface temperature (LST) is compared and analyzed. In this paper, Dali area is taken as the research area, and nine spatio-temporal fusion methods are used to fuse Landsat and MODIS data, and the differences in spatio-temporal simulation effect and calculation efficiency are evaluated through visual discrimination and statistical analysis. Experiments show that: 1) Compared with other spatio-temporal fusion methods, OL-FSDAF2.0 can better restore the real information and structural information of the surface; 2) The computational efficiency of object-level spatiotemporal fusion method is 20.7081 times higher than other pixel-level spatiotemporal fusion methods; 3) Object-level spatio-temporal fusion method has higher ability to capture the details of temporal dynamic features of ground objects than pixel-level spatio-temporal fusion method. Generally speaking, the object-level spatio-temporal fusion method has higher computational efficiency and more accurate fusion effect, among which OL-FSDAF2.0 performs well in complex surface areas and simulated dynamic changes of surface cover..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 51 (2024)
Research on Branch and Foliage Separation Method with Point Transformer v2
Jin MA, Yiping CHEN, Ting HAN, Chaolei WANG, Xiaohai ZHANG, and Wuming ZHANG
Accurate and efficient point cloud branch and foliage separation are essential for accurately calculating above-ground biomass and carbon stocks. However, current methods are computationally expensive and rely on priori knowledge leading to insufficient generalization. To address the above problems, the article proposeAccurate and efficient point cloud branch and foliage separation are essential for accurately calculating above-ground biomass and carbon stocks. However, current methods are computationally expensive and rely on priori knowledge leading to insufficient generalization. To address the above problems, the article proposes to use the point feature-based Transformer network for automated branch and leaf separation study in forest scene 3D point cloud. This research uses the Point Transformer v2 network. Firstly, the grid coding module is used to extract the learnable local structural relations and preserve the geometric topology of the point cloud; secondly, group attention is used to achieve multi-channel joint learning, reduce the redundancy of features and improve the efficiency of computation; finally, a point-based Transformer network is constructed to achieve high-precision semantic segmentation of 3D point clouds of forest trees, and this method is capable of accurate separation of tree trunks and leaves. In this paper, we use the 3D point cloud data of seven different tree species sample plots in Canada and Finland acquired by ground-based laser scanner to conduct branch and foliage separation experiments and accuracy evaluation, and the experimental results show that the OA of the network proposed in this paper is 94.42%, and the mIoU is 78.89%, which indicates that the network in this paper can effectively improve the segmentation accuracy of the tree crown under the conditions of irregular forest tree crown distribution and occlusion. Meanwhile, more accurate segmentation can be achieved for branch and foliage of different tree species and different point cloud densities..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 62 (2024)
Simultaneous Non-Photosynthetic Vegetation and Photosynthetic Vegetation Fractional Cover Estimation Using ZY-1 02D Satellite Imagery
Jia TIAN
In the study area of the forest-grass ecotone in Weichang County, Hebei Province, spectral reflectance data of non-photosynthetic vegetation (NPV), bare soil (BS), and photosynthetic vegetation (PV) were utilized from spectral libraries. Based on the spectral curve shape differences of PV, NPV, and BS in the visible anIn the study area of the forest-grass ecotone in Weichang County, Hebei Province, spectral reflectance data of non-photosynthetic vegetation (NPV), bare soil (BS), and photosynthetic vegetation (PV) were utilized from spectral libraries. Based on the spectral curve shape differences of PV, NPV, and BS in the visible and near-infrared (VNIR) range, two narrow near-infrared spectral bands from the ZY-1 02D satellite's Advanced Hyperspectral Imager (AHSI), as well as two wide near-infrared spectral bands (B8 and B9) from the multispectral camera, Visible Near-Infrared Camera (VNIC), were employed to construct the Normalized Spectral Separation Index (NSSI) for NPV, PV, and BS, respectively. Subsequently, by combining the triangular space constructed based on the Normalized Difference Vegetation Index (NDVI), pixel unmixing was conducted to simultaneously estimate the coverage of NPV, PV, and BS components. Additionally, NSSI was compared with Cellulose Absorption Index (CAI) to evaluate the capability of ZY-1 AHSI in estimating the coverage of non-photosynthetic vegetation in the Shortwave Infrared spectral bands (SWIR). The study demonstrates that the combination of NSSI can be utilized to overcome the influence of the shortwave infrared spectral band, which is easily being affected by water, in calculating non-photosynthetic vegetation coverage, thereby being widely applicable to vegetation remote sensing coverage estimation in both arid and humid regions..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 73 (2024)
Text-Semantics-Driven Feature Extraction from Remote Sensing Imagery
Sijun DONG, and Xiaoliang MENG
With the rapid development of remote sensing technology, high-precision remote sensing image feature extraction has become increasingly crucial in fields such as geographic information science, urban planning, and environmental monitoring. However, traditional image-based remote sensing image feature extraction methodsWith the rapid development of remote sensing technology, high-precision remote sensing image feature extraction has become increasingly crucial in fields such as geographic information science, urban planning, and environmental monitoring. However, traditional image-based remote sensing image feature extraction methods often have limited accuracy when dealing with complex and variable surface features, making it difficult to meet diverse application needs. To address this issue, this paper proposes a novel multimodal remote sensing image semantic segmentation framework (MMRSSEG) that integrates both visual and textual information using deep learning techniques to achieve high-precision analysis of remote sensing images. We conducted a series of experiments on a remote sensing image dataset of buildings, and the results show that MMRSSEG significantly improves the accuracy of pixel-level remote sensing image feature extraction compared to traditional image segmentation methods. In the building recognition task, our method outperformed traditional unimodal algorithms. These experimental results fully demonstrate the effectiveness and prospects of integrating multimodal textual information in remote sensing image segmentation..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 82 (2024)
Water Extraction Method of High Resolution Remote Sensing Image Based on ASPP-SCBAM-DenseUnet
Yuting XIE, Ping LIU, Wenming SHEN, Yu GAO, Shufeng HAO, Xin HAN, and Yuang LI
Aiming at the problems of insufficient attention to detailed information such as small water bodies and water edges in remote sensing image water body extraction research, as well as poor water body connectivity, this paper proposes a densely connected U-shaped network ( ASPP-SCBAM-DenseUnet ) based on improved atrous Aiming at the problems of insufficient attention to detailed information such as small water bodies and water edges in remote sensing image water body extraction research, as well as poor water body connectivity, this paper proposes a densely connected U-shaped network ( ASPP-SCBAM-DenseUnet ) based on improved atrous spatial pyramid pooling and stochastic convolutional block attention module. In this paper, the Dense Block block is used to form the encoder and decoder parts of Unet, and the SCBAM attention mechanism is introduced to reduce noise interference and improve the accuracy of water boundary segmentation. Secondly, the ASPP_SCBAM module is added to set different atrous rates, expand the receptive field, and combine the shallow and deep features of small water bodies to compensate for the feature loss caused by the sampling process. Finally, the network is trained by combining the joint loss function of Dice coefficient and pixel-level binary cross entropy to effectively deal with the unbalanced data set caused by small water bodies. This not only ensures the accuracy of segmentation, but also produces a smoother and more continuous segmentation boundary, thus preventing the model from overfitting or over-refinement. The experimental results show that the scores of pixel accuracy, recall and F1-score extracted by ASPP-SCBAM-DenseUnet network model are 94.19%, 94.29% and 95.15%, respectively, and the scores of frequency weighted intersection over union and mean intersection over union are 89.02% and 88.63%, respectively, which are significantly better than those of semantic segmentation networks such as Unet and Linknet. At the same time, it reduces the misclassification and omission of water bodies, optimizes the edge details of water bodies, and improves the identification of small water bodies and the connectivity of water bodies..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 92 (2024)
Road Extraction Method of High-Resolution Remote Sensing Images Based on Dense Blocks and Improved LinkNet
Zengyou WANG, Xianhua ZHANG, Rong LIU, Zhigao CHEN, and Wanghuang ZHU
Aiming at the problem that feature information is easily lost and lacks attention to target features when the LinkNet network model performs road image segmentation tasks, a high resolution remote sensing image road extraction method based on an improved residual network in LinkNet is proposed. Replace the residual bloAiming at the problem that feature information is easily lost and lacks attention to target features when the LinkNet network model performs road image segmentation tasks, a high resolution remote sensing image road extraction method based on an improved residual network in LinkNet is proposed. Replace the residual block (Res Block) in the coding area of the original LinkNet model with a dense block (Dense Block). The dense connection method reduces the loss of feature information during the transmission process, and builds convolutional attention after each dense block. Units are used to improve the model’s learning ability of target features. Finally, the atrous space pyramid pooling module is used to connect the encoding area and the decoding area to expand the receptive field while also accepting multi-scale target feature information. Experiments show that the accuracy, average intersection ratio and F1-score of this method on the DeepGlobe data set are 82.16%, 83.21% and 81.65%, respectively, which are all better than similar networks. By comparing the extracted road network results, the algorithm has significantly improved the completeness and accuracy of road network extraction under tree shelters and building shadows..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 107 (2024)
Combined Multi-Temporal Sentinel-2 and DeepLabV3+ for County Citrus Information Extraction
Linhai YU, Kaiyao WEI, Shiqing DOU, Bohan DING, and Bing HAN
Aiming at the problem of low accuracy of traditional classification methods in identifying citrus planting spatial information, this paper proposes a county-level citrus planting spatial information extraction method combining multi-temporal Sentinel-2 and DeepLabV3+. Firstly, the optimized lightweight network MobileNeAiming at the problem of low accuracy of traditional classification methods in identifying citrus planting spatial information, this paper proposes a county-level citrus planting spatial information extraction method combining multi-temporal Sentinel-2 and DeepLabV3+. Firstly, the optimized lightweight network MobileNetV2 is used as the backbone network, and the CBAM (Convolutional Block Attention Module) attention mechanism module is embedded to construct the improved lightweight DeepLabV3+; Then, the multi-temporal Sentinel-2 images are used to integrate the original band and spectral index to form the feature data set, and the optimal feature combination and phase of the model classification are determined through experimental comparative analysis; Finally, the image of the study area is segmented into a set of images to be predicted with overlap, and the optimal classification model is used to predict and splice to obtain the results of citrus orchard extraction. The results show that: 1) the extraction accuracy of the improved lightweight DeepLabV3+ is higher than that of DeepLabV3+ and Random Forest model. In the feature combination with the addition of red edge index RESI in the B2-B8A band, the OA can up to 91.1% , and the optimal extraction phase is November. 2) The extraction effect of the overlap prediction method is better than that of the direct prediction method, the edge error of the extracted citrus orchard map spots is basically eliminated, and the relative error between the extracted area and the statistical data of the whole region is kept within ±0.04%, which is of high applicability. This method can provide a reference for the automatic monitoring and planting planning of citrus orchards in the county area in southern China..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 118 (2024)
A Novel Ocean LiDAR Calibration Method Based on Satellite-Borne Passive Remote Sensing
Jinghao ZHANG, Weidong SHANG, Zhenmin SHEN, Song YANG, Tong LI, Xiaomin YE, Kun LIANG, Liang CHENG, Guoqing ZHOU, and Yongchao ZHENG
Ocean Lidar has the technical characteristics of strong water penetration, high vertical resolution and all-day observation, so it has great application potential in satellite ocean profile measurement. But the accurate measurement of ocean Lidar is inseparable from the accurate calibration of system constant. To meet Ocean Lidar has the technical characteristics of strong water penetration, high vertical resolution and all-day observation, so it has great application potential in satellite ocean profile measurement. But the accurate measurement of ocean Lidar is inseparable from the accurate calibration of system constant. To meet this requirement, a novel calibration method for ocean lidar system constant based on satellite-borne passive remote sensing is proposed. This method is based on the combined model of ocean active and passive optical remote sensing. By using the remote sensing reflectivity data obtained by satellite-borne passive remote sensing technology, real-time and efficient calibration of ocean LiDAR system constants can be realized on a global scale. Based on the self-developed airborne ocean Lidar and in-orbit Aqua-MODIS data, the proposed calibration method is validated by flight tests in the Beibu Gulf of Guangxi. The results show that the method is accurate and effective. After the calibration of LiDAR system constant, the root mean square of mean relative difference between the measured value and the theoretical value is 13.7%..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 131 (2024)
Canopy Closure Inversion of Regional Plantation Based on Li-Strahler Geometric-Optical Model
Xinwei YANG, Chunxiang CAO, Min XU, and Kaimin WANG
Forest canopy closure is an important factor in describing forest structure. The traditional remote sensing inversion empirical model of canopy closure has high requirements for ground sample data and has the problem of difficulty in generalizing to large regional scales. In order to solve the above problems, this papeForest canopy closure is an important factor in describing forest structure. The traditional remote sensing inversion empirical model of canopy closure has high requirements for ground sample data and has the problem of difficulty in generalizing to large regional scales. In order to solve the above problems, this paper proposes a regional plantation closure inversion method by combining Li-Strahler geometric-optical model and linear spectral decomposition technique. The principle of this method is to use the average tree height and crown radius of different dominant tree species as geometric-optical model input parameters for each subcompartment, based on the extracted abundance of sunlit background endmember through linear spectral decomposition, then the inversion of canopy closure can be achieved. Based on the proposed method, inversion of the canopy closure of regional plantation was carried out in Chifeng and Nanning study areas, the accuracy of the estimated results was then verified by field measured data. The accuracy evaluation results showed that: the root mean square error and the mean absolute percentage error are 0.08 and 9.45% respectively in the plain region, and 0.12 and 14.04% respectively in the mountainous region for Chifeng study area; meanwhile, the root mean square error and the mean absolute percentage error are 0.11 and 8.68% respectively in the plain region, and 0.13 and 13.27% respectively in the mountainous region for Nanning study area; The results above indicate that the inversion method proposed in this paper can effectively improve the inversion accuracy of canopy closure in regional scale plantations, and has important application value..
Spacecraft Recovery & Remote Sensing
- Publication Date: Jul. 31, 2024
- Vol. 45, Issue 3, 137 (2024)
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20