Highlights

Please wait a minute...


For Selected: Download Citations EndNote Reference Manager ProCite BibTeX RefWorks Toggle Thumbnails

Select

Research on Key Technologies and Applications of Feature Extraction and Knowledge Mining in Planetary Exploration LING Zongcheng, LI Bo, WEI Guangfei, GUO Dijun, LYU Yingbo, LIU Changqing, ZHU Kai, CHEN Jian, ZHAO Qiang, LI Jing, HU Guoping, WANG Jiao, LIU Jianzhong Frontiers of Data and Computing    2025, 7 (4): 3-19.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.001 Abstract （115）   HTML （6）    PDF（pc） （74103KB）（53）       Knowledge map    [Purpose] This article focuses on the demand for deep mining and intelligent feature extraction of planetary exploration data. It investigates key technologies and applications for extracting feature information and knowledge mining from massive remote sensing data obtained during various planetary explorations. [Method] This study breaks through the challenges of reconstructing, fusing, and visualizing multi-source and heterogeneous planetary data, addressing the limitations of insufficient imaging information from single sensor to generate high-quality remote sensing images rich in spatial and spectral information. Spectral characteristic parameters of lunar and martian minerals are extracted from visible near-infrared spectroscopic data, enabling the inversion of the content and distribution of elements and minerals on the lunar and martian surface. Additionally, microwave and radar data are utilized to obtain information regarding the thickness and physical properties of lunar regolith, along with the subsurface structure of typical regions on the Moon and Mars. By employing images and elevation data of the Moon and Mars, multi-scale terrain factors on the surface are calculated, and typical morphological features are automatically extracted using deep learning techniques. [Results] Finally, an integrated platform for displaying and analyzing morphological features, material composition information, and subsurface structures has been established. Furthermore, a planetary data analysis and mining software tool with independent intellectual property rights is developed, which will be released by National Space Science Data Center and Planetary Data System (PDS) Laboratory of Shandong University at Weihai, to support planetary data mapping and geological evolution researches. Table and Figures | Reference | Related Articles | Metrics

Select

Construction of an Intelligent Retrieval System for the Virtual Space Science Observatory LI Yunlong, JIAO Qirong, WANG Cifeng, ZOU Ziming Frontiers of Data and Computing    2025, 7 (4): 20-32.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.002 Abstract （231）   HTML （5）    PDF（pc） （7621KB）（29）       Knowledge map    [Background] With the rapid growth of space science data, the traditional metadata-based retrieval methods have gradually become insufficient to meet the needs of researchers for complex semantic queries. There is an urgent need to introduce intelligent retrieval systems capable of semantic understanding. [Objective] This study aims to develop an intelligent retrieval system for space science data, addressing the limitations of conventional metadata-driven approaches in semantic comprehension and multi-modal data retrieval, thereby enhancing the efficiency and accuracy of accessing heterogeneous space science datasets. [Methods] The proposed system employs a dynamic semantic parsing mechanism based on large language models, combined with hybrid retrieval strategies integrating BM25 and dense vector search methods. For image and time-series data, feature representations are extracted using models such as DINOv2, VISTA, and Timer-XL to construct a multi-modal semantic index. The system adopts a hierarchical architecture that integrates full-text search and vector databases, supporting multiple query modes including natural language, tags, and data examples. [Conclusion] The intelligent retrieval system for the virtual space science observatory significantly enhances the flexibility and accuracy of data discovery by integrating multiple AI models, offering a novel paradigm for the efficient utilization of large-scale space science data. Table and Figures | Reference | Related Articles | Metrics

Select

Automatic Detection and Parameter Extraction of Medium-Scale Traveling Ionospheric Disturbances LAI Chang, LIU Shengyu Frontiers of Data and Computing    2025, 7 (4): 33-41.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.003 Abstract （99）   HTML （0）    PDF（pc） （29629KB）（22）       Knowledge map    [Objective] To address the inefficiency and subjectivity of traditional manual identification for medium-scale traveling ionospheric disturbances (MSTIDs), this study proposes an automatic deep learning-based framework for MSTID detection and parameter extraction from OI airglow images based on the data collected by the all-sky airglow imager at the Meridian Project’s Xinglong station. [Methods] The framework adopts a three-stage architecture: (1) A custom 10-layer convolutional neural network (CNN) is designed to classify raw airglow images, filtering out environmental interference (e.g., clouds, twilight overexposure) and retaining suitable observational data captured at starring night; (2) A Faster Region-based Convolutional Neural Network (Faster R-CNN) model, trained via transfer learning and virtual data augmentation, achieves precise localization of MSTID waveforms; (3) Edge detection and linear fitting algorithms are developed to extract propagation parameters such as direction, velocity, and wavelength. Innovatively, a hybrid Poisson-Gaussian noise model and a waveform simulation function are proposed to generate synthetic training data, enhancing model robustness against limited real-world samples. [Results] Evaluated on the test dataset, the framework demonstrates high performance: the CNN classifier attains 96.9% accuracy in identifying clear-sky conditions; the Faster R-CNN detector achieves an average Intersection-over-Union (IoU) of >75% for wavefront localization. The proposed system significantly improves the automation and objectivity of MSTID analysis, enabling efficient large-scale statistical studies of ionospheric disturbances. Table and Figures | Reference | Related Articles | Metrics

Select

Automated Detection of Geomagnetic Ultra-Low Frequency Waves Based on Geoformer FANG Shaofeng, ZOU Ziming Frontiers of Data and Computing    2025, 7 (4): 42-53.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.004 Abstract （131）   HTML （6）    PDF（pc） （18635KB）（21）       Knowledge map    [Objective] Considering the limitations of traditional methods, such as insufficient classification accuracy of geomagnetic ultra-low frequency (ULF) waves in noisy environments and inferior capability in modeling long-term temporal dependencies, we propose Geoformer, a novel framework integrating depthwise separable convolution and temporal Transformer. By fusing convolutional neural networks (CNNs) with an improved Transformer architecture and optimizing positional encoding strategies, Geoformer enhances the recognition accuracy and model generalization ability for ULF waves. [Methods] This method uses CNNs to extract the local time-domain features and multi-channel spatial correlations of geomagnetic signals. Meanwhile, through time absolute positional encoding (tAPE) and efficient relative positional encoding (eRPE), it strengthens the model’s perception of the absolute positions and relative distances of signal sequences. Finally, it employs the multi-head self-attention mechanism to capture long-term temporal dependencies and multi-dimensional interactive features, achieving accurate classification. [Results] Experimental results indicate that, compared with traditional CNNs and basic Transformer models, Geoformer improves the classification accuracy by 12.2% on real world geomagnetic datasets and is significantly superior to traditional deep learning models like LSTM, GRU, and ResNet. [Limitations] The model’s computational complexity grows quadratically with the increase of signal length, necessitating the incorporation of downsampling techniques in real-time processing of ultra-long time sequences. Moreover, it relies on high-quality labeled data, so transfer learning or self-supervised pre-training should be introduced in small sample scenarios. [Conclusions] Through the innovation of the CNN-Transformer architecture and the optimization of positional encoding, this paper provides an efficient solution for the intelligent analysis of geomagnetic ultra-low frequency wave signals, holding significant promise for playing an important role in fields such as geophysical monitoring and space weather early warning. Table and Figures | Reference | Related Articles | Metrics

Select

Software Development for Martian Mineral Identification and Distribution Mapping LIU Changqing, LYU Yingbo, LING Zongcheng Frontiers of Data and Computing    2025, 7 (4): 54-66.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.005 Abstract （68）   HTML （4）    PDF（pc） （13692KB）（18）       Knowledge map    [Objective] Mars exploration missions have acquired massive imaging spectral data, presenting significant challenges to traditional manual processing and visual interpretation approaches. This study introduces a generalized methodology for processing Martian imaging spectral data, coupled with the development of a software for automatic Martian mineral identification and distribution mapping. [Methods] The software leverages the FATT algorithm based on RELAB spectral datasets to enable automated mineral type classification, integrates spectral matching techniques (SAM, SID, SID_SA) for generating spatial distribution maps, and derives spectral parameter distributions by analyzing visible-near-infrared features (e.g., absorption depth) of Martian minerals. [Results] The validation using preprocessed imaging spectral data from MMS, CRISM, and OMEGA payloads demonstrates that the software facilitates efficient mineral identification and distribution mapping, substantially streamlining processing workflows and reducing analysis time for planetary scientists. Table and Figures | Reference | Related Articles | Metrics

Select

Research on Construction of a Semantic Association-Driven Space Science Data Repository System and Dataset Association Recommendation WU Zhaochen, LU Changfa, LI Gang, LAN Chenyang, WANG Cifeng Frontiers of Data and Computing    2025, 7 (4): 67-78.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.006 Abstract （112）   HTML （5）    PDF（pc） （16097KB）（14）       Knowledge map    [Background] With the exponential growth of multimodal data in space science, existing data management systems face significant challenges. The lack of semantic correlations between data in traditional architectures severely limits the efficiency of interdisciplinary knowledge discovery. [Objective] This study aims to construct a semantically enhanced space science data repository system, deeply exploring metadata semantics and their correlations across multi-source data to break disciplinary barriers and enhance correlation analysis capabilities. [Methods] The research constructs a metadata semantics network for space science data through a progressive three-tier conceptual-logical-physical architecture. Employing a non-intrusive data integration methodology, we develop key components including archival repository interface services, unified external service APIs, graph database management systems, and graph query engines, thereby establishing the space science data repository system without modifying existing business architectures. Furthermore, we design a metadata-driven semantic similarity calculation algorithm to quantify the association strength between datasets, with technical validation conducted through related datasets recommendation experiments. [Conclusions] Experiments show that the proposed method effectively improves knowledge discovery efficiency in space science, offers a novel solution to multimodal data fusion challenges, and significantly enhances capabilities for analyzing complex scientific data. Table and Figures | Reference | Related Articles | Metrics

Select

Extracting the Magnetopause Structure Based on Direct Volume Visualization Methods ZHONG Jia, ZOU Ziming Frontiers of Data and Computing    2025, 7 (4): 79-88.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.007 Abstract （67）   HTML （2）    PDF（pc） （41434KB）（16）       Knowledge map    [Objective] Visualization is an important way to effectively analyze the magnetosphere configuration, and it is of great significance to the understanding of the energy exchange mechanism between solar wind and magnetosphere. [Methods] The boundary points should be located and then triangular facets are constructed before drawing the three-dimensional magnetopause based on the streamline method while the interior structure of the magnetopause is invisible. This paper attempts to introduce a direct volume visualization method without locating boundary points to perspectively display the magnetopause structure, and designs transfer functions based on the value rate distribution map and k-means++ algorithm. [Results] The experiment utilized five features, including plasma density gradient, magnetic induction intensity gradient, and so on from PPMLR-MHD simulation data, and successfully reconstructed the complete three-dimensional structures of the magnetopause and bow shock. Among these, the visualization of the magnetopause based on solar wind density gradient and magnetic induction intensity gradient yields clearer results. [Conclusions] The results demonstrate that the magnetopause direct volume visualization method proposed in this study offers more robust information and spatial representation capabilities compared to geometric rendering approaches, providing valuable insights for subsequent visualization of other structures within the magnetosphere. Table and Figures | Reference | Related Articles | Metrics

Select

A Multi-Dimensional Clustering Method for Morphological Characterization of Lunar Impact Craters LIU Fangchao, ZHANG Li, GUO Dijun, CHEN Jian, LYU Yingbo, LING Zongcheng, LI Boran, LI Xinyu, MA Yunlong Frontiers of Data and Computing    2025, 7 (4): 89-100.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.008 Abstract （72）   HTML （2）    PDF（pc） （6802KB）（12）       Knowledge map    [Purpose] Lunar impact craters are important landmarks for studying the geologic evolution of the lunar surface, while systematically classifying and morphologically analyzing impact craters with the support of high-resolution image data is still an important research topic. [Method] In this study, based on the LROC NAC image data near the Chang'e-3 landing site, a k-means clustering method without training is proposed for automatic classification of impact craters with similar morphological features. Firstly, feature information in five dimensions, including mean, variance, entropy, contrast, and scale, is extracted from each image. Secondly, the optimal number of clusters, k=8, is determined by four methods, namely, the elbow method, the silhouette coefficient method, the Gap statistic, and hierarchical clustering. Thirdly, the feature information and the k are input into the k-means algorithm for clustering of the impact crater images. Finally, the images are clustered by drawing a cluster feature scatter plot and example images, etc. to visualize and analyze the clustering results. [Results] Based on the clustering results, the morphological features of different types of impact craters are analyzed in detail, revealing the potential significance of different morphological categories of impact craters in the geological evolution of the lunar surface. [Conclusion] This study provides a new method for the automated classification and morphological characterization of lunar impact craters, which provides important support for an in-depth understanding of lunar surface geological processes. Table and Figures | Reference | Related Articles | Metrics

Select

Scientific Methods for Analyzing Observational Data of Gamma-Ray Bursts and Related High-Energy Transients YANG Jun, ZHANG Binbin Frontiers of Data and Computing    2025, 7 (4): 101-117.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.009 Abstract （76）   HTML （4）    PDF（pc） （3290KB）（10）       Knowledge map    [Context] Astronomical research heavily relies on the acquisition and analysis of observational data. For high-energy transients such as gamma-ray bursts (GRBs), radiation undergoes rapid and intense variations over extremely short timescales, characterized by significant temporal and spectral evolution. Efficiently and accurately extracting these dynamic features from complex observational data has become a critical technical challenge in high-energy time-domain astrophysics. [Objective] This study aims to develop a scientific data analysis pipeline tailored for GRBs and related high-energy transient sources, with a particular focus on the development and application of key algorithms and core technologies that ensure high efficiency and precision. [Methods] For temporal analysis, we propose an automated signal identification and background fitting algorithm that integrates baseline correction, Bayesian block segmentation, significance calculation, and polynomial fitting. For spectral analysis, we construct a Bayesian inference-based spectral fitting framework designed to robustly estimate model parameters and their uncertainties. [Results] The proposed analysis pipeline and its key algorithms have been successfully applied to real observational data of GRBs. The approach effectively identifies and separates signal from background and enables reliable estimation of key spectral parameters through a Bayesian inference algorithm. [Conclusion] The data analysis framework presented in this work significantly improves the efficiency and accuracy of observational data analysis for GRBs and related high-energy transient sources. It offers a reusable technical approach for the automated analysis of high-energy astrophysical explosion phenomena and holds substantial potential for broad application and scientific advancement. Table and Figures | Reference | Related Articles | Metrics

Select

Research on Joint Object Detection Method for Multi-Band Images of Space Hurricane SHI Ke, LU Yang, LU Sheng, WANG Yong, ZOU Ziming Frontiers of Data and Computing    2025, 7 (4): 118-128.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.010 Abstract （73）   HTML （3）    PDF（pc） （33682KB）（12）       Knowledge map    [Objective] As a typical phenomenon triggered by solar-terrestrial interactions, space hurricane often generates a huge energy cyclone in the middle and upper atmosphere, which is accompanied by the occurrence of auroral phenomena. Identifying auroral images can assist scientists in finding typical space hurricane events. However, currently, the search for such events mainly relies on experts' manual identification of auroral images, which is rather inefficient. To solve the above problems, this study explores a deep learning-based joint object detection method for multi-band images, achieving event recognition and precise localization of space hurricanes. [Methods] In this study, four-band auroral images (121.6nm, 135.6nm, LBHS, LBHL) from DMSP/SSUSI are used to identify space hurricane events. Based on the YOLOv8 algorithm framework, target-level fusion and feature-level fusion strategies are introduced. Meanwhile, single-band and multi-band fusion models for space hurricane recognition are established. [Results] In the event recognition task, by comparing the experimental results of single-band baseline models and multi-band fusion models, it is shown that the 1216_LBHL combination in feature-level fusion performs best, with an F1 score of 0.941. In the object detection task, the 1216_LBHL combination in target-level fusion achieves the highest AP value of 0.917. [Conclusions] Feature-level fusion demonstrates greater advantages in space hurricane event recognition, while target-level fusion is more suitable for object detection tasks. This indicates that the combined optimization of multi-band complementarity and fusion strategies is the key to enhancing detection performance. Table and Figures | Reference | Related Articles | Metrics

Select

Lunar Rock Thin Section Image Classification in Label-Scarce Scenarios via DINO-Based Feature Transfer DAI Minhao, DONG Junfeng, CHEN Jian, LYU Yingbo, ZHANG Li, LING Zongcheng Frontiers of Data and Computing    2025, 7 (4): 129-142.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.011 Abstract （72）   HTML （3）    PDF（pc） （25072KB）（12）       Knowledge map    [Background] Lunar rock thin-section images contain rich information about geological evolution. However, due to limited sample availability, dataset imbalance, and high annotation costs, traditional supervised learning-based classification methods face significant application challenges. [Methods] To address this, this paper proposes a self-supervised contrastive learning framework based on the DINO model, which extracts image features and integrates them with various classifiers to enable automatic recognition and analysis without requiring labeled data. A lunar rock image dataset was constructed, and contrastive learning was employed for feature modeling, followed by evaluation using multiple classifiers. [Results] Experimental results demonstrate that features extracted by the self-supervised model achieve outstanding performance with classifiers such as KNN and MLP, reaching a maximum classification accuracy of 91.56% from 45.11%, and it did not exhibit the problem of class imbalance even when the sample sizes were significantly different. The t-SNE visualization and confusion matrix analysis further confirm the model's effectiveness in feature clustering and category discrimination. The model exhibits strong robustness and generalization capabilities. [Conclusion] This study provides a feasible approach for automated interpretation of lunar rock images, supporting research on lunar geological evolution and related deep-space exploration missions. Table and Figures | Reference | Related Articles | Metrics

Select

Research on Anomaly Detection Algorithms for Scientific Satellite Images Based on PatchCore WANG Lei, MA Fuli, YU Qinsi, WEI Mingyue Frontiers of Data and Computing    2025, 7 (4): 143-154.   DOI: 10.11871/jfdc.issn.2096-742X.2025.04.012 Abstract （80）   HTML （6）    PDF（pc） （12480KB）（27）       Knowledge map    [Objective] With the rapid growth of observational data from space science satellites, image anomaly detection has become a critical component for ensuring data quality and supporting scientific research, which requires urgently the development of efficient and automated methods. [Context] Due to the scarcity or even absence of anomaly samples during the early stages of satellite operation, traditional supervised learning approaches are difficult to apply directly. Therefore, this study proposes an anomaly detection framework tailored for space science satellite images based on unsupervised learning. [Methods] Based on the PatchCore algorithm, the proposed method designs modules for feature extraction, core-set construction, anomaly scoring, and image classification. In addition, multiple anomaly scoring and threshold setting strategies based on statistical analysis and clustering methods are explored to enhance detection sensitivity and robustness. [Results] Extensive experiments were conducted on actual solar observation datasets, with comparative analysis against mainstream unsupervised detection methods such as PaDiM and CS-Flow. The results demonstrate that the proposed method achieves outstanding performance, with AUROC and AUPR values reaching to 0.9996 and 0.9999, respectively. In terms of system implementation, ONNX Runtime is adopted for lightweight model deployment, which significantly improves inference speed and deployment flexibility, and establishes a complete closed-loop process covering data acquisition, anomaly detection, and alert feedback. [Conclusions] The study shows that the developed system can effectively enhance the efficiency of image quality monitoring in space science missions, providing valuable references for the construction of intelligent anomaly detection systems in future space observation tasks. Table and Figures | Reference | Related Articles | Metrics

Select

A Review of Research on Social Network Influence Prediction Based on Multi-Class Features SHUI Yingyi, ZHANG Qi, LI Gen, ZHANG Shihao, WU Shang Frontiers of Data and Computing    2025, 7 (1): 2-18.   DOI: 10.11871/jfdc.issn.2096-742X.2025.01.001 Abstract （385）   HTML （33）    PDF（pc） （2048KB）（801）       Knowledge map    [Objective] Influence prediction, as an important content of social network analysis, has important social value and practical significance in many fields such as public opinion monitoring, online marketing, intelligence analysis, personalized recommendation, advertisement positioning, and communication prediction. Early influence prediction methods based on feature engineering established the relationship between different features and popularity by extracting and constructing key features. This paper focuses on the multi-class features related to social network influence, and conducts research and review from the aspects of multi-class feature extraction, prediction model construction, and prediction evaluation methods, aiming to comprehensively analyze the existing research methods, and provide reference for improving the accuracy of social network influence prediction. [Methods] Based on the current widely adopted deep learning methods, this paper summarizes and elaborates on the visual, textual, emotional, temporal, and user features of social networks by reviewing the literature, and analyzes the current research status and limitations of the influence prediction methods of social networks based on multi-class features. [Conclusions] With the development of deep learning theory, breakthrough progress has been made in deep feature extraction and prediction model construction, but at present, in terms of social network influence prediction, feature combination prediction methods based on multi-class features are still insufficient, and it is necessary to study more effective feature pre-extraction models to improve social network influence prediction accuracy. Table and Figures | Reference | Related Articles | Metrics

Select

Review of Research on Chart Question Answering MA Qiuping, ZHANG Qi, ZHAO Xiaofan Frontiers of Data and Computing    2025, 7 (1): 19-37.   DOI: 10.11871/jfdc.issn.2096-742X.2025.01.002 Abstract （295）   HTML （24）    PDF（pc） （4493KB）（671）       Knowledge map    [Objective] The purpose of this paper is to comprehensively review the research progress of Chart Question Answering (CQA) technology, analyze existing models and methods, and explore future development directions. [Methods] Firstly, CQA models are divided into two categories: deep learning-based and multi-modal large models. Deep learning-based methods are further subdivided into end-to-end models and two-stage models in this paper. Subsequently, the three core processes taken by the deep learning-based CQA are deeply analyzed, and a detailed classification along with an in-depth analysis of the existing processing methods for each process is provided. CQA models based on multi-modal large models are also explored in this paper, with their advantages, limitations, and future development directions being analyzed. [Results] The current research status of CQA technology is comprehensively summarized, and an in-depth analysis of existing models and methods is conducted. It is found that deep learning-based CQA models perform well in handling standard chart types and simple tasks, but fall short when facing complex, non-standardized charts or tasks requiring deep reasoning. In contrast, CQA models based on multi-modal large models show great potential, but the improvement in model performance often comes with an increase in model size and computational complexity. Future research should focus on developing more lightweight question answering models and enhancing model interpretability. Table and Figures | Reference | Related Articles | Metrics

Select

A Survey of Face Age Editing Based on Generative Adversarial Networks and Diffusion Models JIN Jiali, GAO Siyuan, GAO Manda, WANG Wenbin, LIU Shaozhen, SUN Zhenan Frontiers of Data and Computing    2025, 7 (1): 38-55.   DOI: 10.11871/jfdc.issn.2096-742X.2025.01.003 Abstract （416）   HTML （17）    PDF（pc） （21416KB）（133）       Knowledge map    [Purpose] In recent years, deep generative models have made significant progress in the task of facial age editing. This paper summarizes facial age editing methods based on deep generative models such as Generative Adversarial Networks (GANs) and diffusion models. [Methods] This survey first introduces the basic concepts of face age editing, relevant datasets, and evaluation metrics. It then analyzes the applications of commonly used GANs, Diffusion Models, and their variants in age editing tasks. The performance of existing models in terms of age accuracy, identity consistency, and image quality is summarized, and the suitability of different evaluation metrics is discussed. [Results] Age editing technology based on GANs and Diffusion Models have achieved significant improvements in image quality and age prediction accuracy. However, challenges remain in generating fine details, particularly when dealing with large age gaps. [Conclusions] Future research in face age editing can further enhance model generation capability and application effects by developing larger, higher-quality datasets and integrating 3D face reconstruction technology with efficient sampling algorithms from Diffusion Models. Table and Figures | Reference | Related Articles | Metrics