Knowledge-Driven GeoAI: Integrating Spatial Knowledge into Multi-Scale Deep Learning for Mars Crater Detection
Open accessThe DOI matches its Crossref record; 75 of the 75 references it cites were found in a registry or an index.
Abstract
This paper introduces a new GeoAI solution to support automated mapping of global craters on the Mars surface. Traditional crater detection algorithms suffer from the limitation of working only in a semiautomated or multi-stage manner, and most were developed to handle a specific dataset in a small subarea of Mars’ surface, hindering their transferability for global crater detection. As an alternative, we propose a GeoAI solution based on deep learning to tackle this problem effectively. Three innovative features are integrated into our object detection pipeline: (1) a feature pyramid network is leveraged to generate feature maps with rich semantics across multiple object scales; (2) prior geospatial knowledge based on the Hough transform is integrated to enable more accurate localization of potential craters; and (3) a scale-aware classifier is adopted to increase the prediction accuracy of both large and small crater instances. The results show that the proposed strategies bring a significant increase in crater detection performance than the popular Faster R-CNN model. The integration of geospatial domain knowledge into the data-driven analytics moves GeoAI research up to the next level to enable knowledge-driven GeoAI. This research can be applied to a wide variety of object detection and image analysis tasks.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures






Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Through data-driven analysis, deep learning models have demonstrated the capability to recognize distinctive flood patterns and subtle flood boundaries, even in regions under complex and rapidly changing conditions.”
From Assessment of a new GeoAI foundation model for flood inundation mapping (2023) · cites this paper as [14]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 2×
“ResNet50 (pretrained on ImageNet) is used as the feature extractor for both the Faster-RCNN and our proposed models.”
BackgroundU-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
Histograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 1×
“This simple yet effective approach has been widely used in both early machine learning methods [52] and recent deep neural networks [53,54].”
Method(2016) · Cited by 31,590
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019) · arXiv · doi:10.48550/arxiv.1912.01703 · Cited by 31,473 · cited 1×
“The models were developed with the PyTorch [77] machine learning framework.”
MethodFeature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939 · cited 3×
“However, even the state-of-the-art multi-scale CNN-based detectors [59,64] cannot be directly applied for crater detection.”
BackgroundHe, K., Gkioxari, G., Dollar, P., & Girshick, R. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.322 · Cited by 29,137
Research works printed but not linked: 2
- Schaber, G.G., and Chadwick, D.J. (1993, January 15–19). Venus’ impact-crater database: Update to approximately 98 percent of the planet’s surface. Proceedings of the Lunar and Planetary Science Conference XXIV, Houston, TX, USA.
- Tewari, A., Verma, V., Srivastava, P., Jain, V., and Khanna, N. (2020). Automated Crater Detection from Co-registered Optical Images, Elevation Maps and Slope Maps using Deep Learning. arXiv.
Cited by
Li, Z. (2022) · Computers, Environment and Urban Systems · doi:10.1016/j.compenvurbsys.2022.101845 · Cited by 1,195
Geomorphometry and terrain analysis: data, methods, platforms and applications
Xiong, L., Li, S., Tang, G., & Strobl, J. (2022) · Earth-Science Reviews · doi:10.1016/j.earscirev.2022.104191 · Cited by 160
Li, W., & Hsu, C.-Y. (2022) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi11070385 · Cited by 147
Self-citationMSA R-CNN: A comprehensive approach to remote sensing object detection and scene understanding
Sagar, A. S. M. S., Chen, Y., Xie, Y., & Kim, H. S. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.122788 · Cited by 79
GeoAI for Science and the Science of GeoAI
Li, W., Arundel, S. T., Gao, S., Goodchild, M. F., Hu, Y., Wang, S., & Zipf, A. (2024) · Journal of Spatial Information Science · doi:10.5311/josis.2024.29.349 · Cited by 93
“The leverage and fusion of multi-source geospatial data, as well as the integration of geospatial domain knowledge, are unique to GeoAI research [19].”
Self-citationWang, S., & Li, W. (2021) · Computers, Environment and Urban Systems · doi:10.1016/j.compenvurbsys.2021.101715 · Cited by 59
Self-citationGeoscience-aware deep learning: A new paradigm for remote sensing
Ge, Y., Zhang, X., Atkinson, P. M., Stein, A., & Li, L. (2022) · Science of Remote Sensing · doi:10.1016/j.srs.2022.100047 · Cited by 64
High-Resolution Feature Pyramid Network for Automatic Crater Detection on Mars
Yang, S., & Cai, Z. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3104925 · Cited by 48
“Similarly, Hsu et al. [15] proposed a GeoAI solution based on deep learning to tackle the problem that traditional crater detection algorithms cannot transfer their knowledge from specific dataset to global crater detection.”
BackgroundHsu, C.-Y., Li, W., & Wang, S. (2024) · International Journal of Geographical Information Systems · doi:10.1080/13658816.2024.2397441 · Cited by 49
Self-citationRezvani, S. M. H. S., Gonçalves, A., Silva, M. J. F., & Almeida, N. M. (2024) · Sustainable Cities and Society · doi:10.1016/j.scs.2024.105873 · Cited by 43
Tewari, A., Verma, V., Srivastava, P., Jain, V., & Khanna, N. (2022) · Planetary and Space Science · doi:10.1016/j.pss.2022.105500 · Cited by 45
Assessment of a new GeoAI foundation model for flood inundation mapping
Li, W., Lee, H., Wang, S., Hsu, C.-Y., & Arundel, S. T. (2023) · Proceedings of the 6th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery · doi:10.1145/3615886.3627747 · Cited by 39
Self-citationYOLO-Crater Model for Small Crater Detection
Mu, L., Xian, L., Li, L., Liu, G., Chen, M., & Zhang, W. (2023) · Remote Sensing · doi:10.3390/rs15205040 · Cited by 34
Li, W., Wang, S., Arundel, S. T., & Hsu, C.-Y. (2022) · GeoInformatica · doi:10.1007/s10707-022-00476-z · Cited by 28
“Scale research has focused on multi-scale analysis to extract scale-dependent features for classification and object detection, and for small object detection leveraging image pyramids and feature fusion [28, 29, 30].”
BackgroundSelf-citationGeoAI Reproducibility and Replicability: A Computational and Spatial Perspective
Li, W., Hsu, C.-Y., Wang, S., & Kedron, P. (2024) · Annals of the American Association of Geographers · doi:10.1080/24694452.2024.2373787 · Cited by 33
“To address this challenge, we ensured that in our data partitioning, each image scene covers the same area, and distortion was corrected algorith-mically before an image scene was used for training (Hsu et al., 2021).”
InfluentialSelf-citationLi, S., Li, K., Xiong, L., & Tang, G. (2022) · Remote Sensing · doi:10.3390/rs14051166 · Cited by 19
Squeezing Data from a Rock: Machine Learning for Martian Science
Nagle-McNaughton, T., Scuderi, L. A., & Erickson, N. (2022) · Geosciences · doi:10.3390/geosciences12060248 · Cited by 21
A Study on the Autonomous Detection of Impact Craters
Aburaed, N., Al-Saad, M., Mansoori, S. A., & Al-Ahmad, H. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20650-4_15 · Cited by 19
Liu, S., Song, M., Chen, S., Fu, X., Zheng, S., Hu, W. D., Gao, S., & Cheng, K. (2023) · Frontiers in Marine Science · doi:10.3389/fmars.2023.1134418 · Cited by 16
“It’s also our plan to open source the modeling code to benefit to the broad research communities and GeoAI (Hsu et al., 2021) technologies will be explored to speedup the computation of location selection of buoy deployments.”
MethodImpact crater recognition methods: A review
Chen, D., Hu, F., Zhang, L., Wu, Y., Du, J., & Peethambaran, J. (2024) · Science China Earth Sciences · doi:10.1007/s11430-023-1284-9 · Cited by 14
“In another work, Hsu et al. (2021) successfully achieved data fusion at the feature fusion level in their impact crater detection network.”
High-Accuracy Crater Detection for Complex Planetary Surfaces
Zhuo, M., Wang, Y., Zhang, X., & Chen, Z. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3560283 · Cited by 15
“Hsu et al. [37] improved the classic Faster R-CNN model and successfully applied it to impact crater detection using the THEMIS IR data, achieving a precision of 84.50% and a recall of 86.62%.”
InfluentialLightweight tensorial convolutional neural network for lunar impact crater detection
Qiu, Y., Liang, Y., Chen, X., Zhang, Z. J., Xie, S., & Zhou, G. (2024) · Advances in Space Research · doi:10.1016/j.asr.2024.03.028 · Cited by 11
Ye, P., Huang, R., Xu, Y., Li, W., Ye, Z., & Tong, X. (2024) · Icarus · doi:10.1016/j.icarus.2024.116358 · Cited by 13
Self-citationSinha, M., & Paul, S. (2025) · Results in Earth Sciences · doi:10.1016/j.rines.2025.100094 · Cited by 8
A Multiscale Vision Transformer-Based Multimodal GeoAI Model for Mapping Arctic Permafrost Thaw
Li, W., Hsu, C.-Y., Wang, S., Gu, Z., Yang, Y., Rogers, B. M., & Liljedahl, A. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3564310 · Cited by 14
“This .”
Self-citationKnowledge-driven deep learning approaches for computer vision tasks: A survey
Benkirane, F. E., Crombez, N., Hilaire, V., & Ruichek, Y. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114645 · Cited by 6
Comparative Planetology – Comparing Cirques on Mars and Earth Using a Cnn
Williams, J. M., Scuderi, L. A., McClanahan, T. P., Banks, M. E., & Baker, D. M. H. (2023) · Geomorphology · doi:10.1016/j.geomorph.2023.108881 · Cited by 6
Boundary Delineator for Martian Crater Instances with Geographic Information and Deep Learning
Liu, D., Cheng, W., Qian, Z., Deng, J., Liu, J., & Wang, X. (2023) · Remote Sensing · doi:10.3390/rs15164036 · Cited by 7
Liu, D., & Cheng, W. (2023) · Frontiers in Astronomy and Space Sciences · doi:10.3389/fspas.2023.1275516 · Cited by 6
Wang, J., Cao, Z., Fu, S., Kang, Z., & Wang, J. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3597739 · Cited by 5
“[14]combinedfeaturepyramidnetworksandscale-awareclassificationinadeep-learning-basedGeoAImodel,im-provingMarscraterclassificationaccuracyby13%comparedto SVMsandrule-basedmethods.”
Liu, Y., Song, T., Xian, C., Chen, R., Zhao, Y. G., Li, R., & Guo, T. (2024) · Remote Sensing · doi:10.3390/rs16112024 · Cited by 4
“Crater detection is also gradually gaining attention [5–9].”
BackgroundMethodInfluentialMemduhoğlu, A., & Başaraner, M. (2023) · Environment and Planning B Urban Analytics and City Science · doi:10.1177/23998083231206165 · Cited by 4
“These technologies facilitate the seamless integration of heterogeneous spatial data sources and increase expressiveness by capturing complex relationships and semantics (Homburg et al., 2016; Hsu et al., 2021; Zhang et al., 2015).”
BackgroundAn introduction to Bayesian geospatial analysis using a Bayesian multilevel model case study
Tu, W., Yu, L., & Tu, J. (2025) · Annals of GIS · doi:10.1080/19475683.2025.2451230 · Cited by 4
“…of modelling results, and informs decision-making about complex spatial processes, for thess reasons, it has been increasingly adopted in big data-driven geospatial science (Carlos et al. 2022; Hsu, Li, and Wang 2021; Kumari, Jyoti, and Dipti 2023; Shen, Choi, and Chan 2019; Vuong et al. 2020).”
A New Lunar Lineament Extraction Method Based on Improved UNet++ and YOLOv5
Yan, P., Liang, J., Tian, X., & Zhai, Y. (2024) · Sensors · doi:10.3390/s24072256 · Cited by 3
Yang, C., Zhao, M., Bruzzone, L., Guan, R., & Zhao, H. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3543861 · Cited by 3
“Object detection is a crucial computer vision task that makes the machine identify and localize objects in images [26], [27], [28], [29], [30].”
Zheng, Q., Huang, R., Xu, Y., Zhang, F., Xiao, C., Li, L., & Tong, X. (2025) · Planetary and Space Science · doi:10.1016/j.pss.2025.106104 · Cited by 3
Evaluation of Resource-Efficient Crater Detectors on Embedded Systems
Vellas, S., Psomas, B., Karadima, K., Danopoulos, D., Paterakis, A., Lentaris, G., Soudris, D., & Karantzalos, K. G. (2024) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss53475.2024.10642518 · Cited by 3
“We employ a large-scale publicly available image dataset of Mars craters [3], consisting of 102,675 images sourced from a global mosaic [68].”
BackgroundMethodInfluentialWambugu, N. M., Wang, R., Shi, T., Chan, T. N., Fei, J., Zhang, Y., Zhou, J., Jia, Z., & Guo, B. (2026) · ISPRS Open Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.ophoto.2026.100128 · Cited by 2
Automated Earthwork Detection Using Topological Persistence
Lapides, D. A., Grindstaff, G., & Nichols, M. H. (2024) · Water Resources Research · doi:10.1029/2023wr035990 · Cited by 2
“Machine learning has been recently introduced for feature detection with great success (e.g., Hsu et al., 2021; Steinfeld et al., 2013; Torres et al., 2020; Wang & Li, 2021; with true positive detection rates of about 60%–98%); data preparation for model training, though, can be time-consuming and…”
BackgroundImpact of artificial intelligence (AI) in Martian architecture (exterior and interior)
Bande, L. (2024) · Materials research proceedings · doi:10.21741/9781644903216-31 · Cited by 1
Multi-scale feature pyramid-based crater detection on lunar surface
Chaini, C., Jha, V. K., & Rajnish, K. (2025) · Earth Science Informatics · doi:10.1007/s12145-025-01818-9 · Cited by 1
“For a fair assessment, this study implemented a similar baseline architecture to train the fusion of datasets using FPN-based ResNet50 as the backbone with Faster RCNN head (Hsu et al. 2021 ) and subsequently prepared Yolo series such as versions 5 and 7 (Zhu et al. 2023; Lin et al. 2017).”
InfluentialContinual Crater Detection With Similarity Distillation
Yang, S., Guo, Y., & Cai, Z. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3601399 · Cited by 1
“Similarly, Hsu et al. [10] proposed a GeoAI solution based on deep learning to tackle the problem of transferring knowledge from a specific dataset to global crater detection.”
CS-YOLO: A cross-scale model designed for impact crater detection in Mars remote sensing images
Xiao, H., Lin, S. Y., & Chen, L. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227269 · Cited by 1
Li, M., Li, J., Wang, Y., & Liu, Y. (2026) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2026.3693662 · Cited by 1
“Contemporary architectures like GACraterNet [7], GeoAI [8], and MarsMapNet [9] excel at object-level detection.”
Chen, X., Zhou, G., Huang, Z., & Xie, S. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3478807
“To name an instance, [12] used faster R-CNN with pyramid layers to learn multiscale DOM representations.”
MethodAn Improved YOLOv11-Based Algorithm for Impact Crater Detection
Zhang, T., Liang, J., & Tian, X. (2025) · IEEE International Conference on Image Processing and Computer Applications (ICIPCA) · doi:10.1109/icipca65645.2025.11138997
Bacha, M. Z., Puppio, M. L., Zucca, M., Staglianò, A., & Sassu, M. (2025) · Lecture notes in civil engineering · doi:10.1007/978-3-032-04350-4_51
A comparison of single-stage and two-stage based crater detectors on the lunar surface
Chaini, C., Jha, V. K., & Rajnish, K. (2026) · Earth Science Informatics · doi:10.1007/s12145-025-02066-7
Li, J., Wang, J., Jia, J., Cheng, W., & Guo, D. (2026) · Journal of Geophysical Research Machine Learning and Computation · doi:10.1029/2025jh000671
Sandeep, V., Niraj, M., & Mohanty, S. K. (2025) · International Conference on Mathematics, Actuarial Science, Computer Science and Statistics (MACS) · doi:10.1109/macs68476.2025.11453442
“It outperforms Faster R-CNN [6] but it can struggle with complex terrains and adaptability.”
Few-Shot Semantic Segmentation Method for Crater Detection on Small Bodies
Li, S. L., & Zhu, S. (2026) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2026-99-1852
Pei, Y., & Wang, P. (2026) · Remote Sensing · doi:10.3390/rs18142395
Computer vision for mars exploration: a review of orbital remote sensing and in situ perception
Jie, C. H. E. N., Liao, X., Jiang, M., Yutong, X. I. A. O., & Yu-Yan Sara, Z. H. A. O. (2026) · Planet · doi:10.15302/planet.2026.26015
Metadata
- Authors
- Hsu, C.-Y., Li, W., & Wang, S.
- Affiliations
- Arizona State University (US)
- Corresponding author
- Li, Wenwen
- Venue
- Remote Sensing
- Publisher
- MDPI AG
- Published
- 2021-05-28
- Volume
- 13
- Issue
- 11
- Pages
- 2116
- ISSN
- 2072-4292 (ISSN-L)
- OpenAlex
- W3172599549
- Cited by
- 53
- References
- 70
- Citations by year
- 2026: 7 · 2025: 14 · 2024: 12 · 2023: 9 · 2022: 9 · 2021: 2
- FWCI
- 2.88
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Remote Sensing (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Science Foundation (1937908, 2033521, 1455349, 1936677, 1853864)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Earth and Planetary Sciences › Space and Planetary Science › Planetary Science and Exploration
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Remote-Sensing Image Classification
- Keywords
- GeoAI, multi-scale object detection, feature pyramid network, Hough transform, Faster R-CNN
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.