Conducting a geological field survey at the initial stage is an important step in geo-oriented projects and construction. Therefore, better and more accurate solutions are only possible with field analysis and proper modeling. The geological modeling process takes a long time, especially depending on the area of interest. It is inefficient to digitize 2D geological maps with traditional software that uses manual user interaction. This paper proposes a state-of-the-art feature detection methodology for detecting geological features on high-resolution maps. With the development of efficient deep learning algorithms and the improvement of hardware systems, the accuracy of detecting specific objects in digital images, such as human facial features, has reached more than 90%. Current object detection models based on convolutional neural networks cannot be directly applied to high-resolution geological maps due to the input image size limitations of conventional object detection solutions, mostly limited by hardware resources. This paper proposes a sliding window method for character detection of geological features. Detection models are trained using transfer learning with You Look Only Once-v3 (YOLO-v3), Single Shot Multi-Box Detector (SSD), Faster-Region-based Convolutional Neural Network (Faster-RCNN), and Single Shot Multi-Box Detector_RetinaNet (SSD_RetinaNet). All models provide competitive success rates with an average precision (AP) of 0.96 on YOLOv3, 0.88 AP on EfficientNet, 0.92 AP on Faster- RCNN, and 0.97 AP on SSD_RetinaNet. YOLOv3 outperformed the best detection over SSD according to F1 recall and score. Since the input size of detection models is limited, a sliding window algorithm is used to separate high-resolution map images. The final detected strike features are provided as a digital dataset that can be used for further manipulations. Thus, Convolutional Neural Network (CNN) based object detection along with a sliding window protocol can be applied to manual map digitization processes to provide instantaneous digitized data with higher accuracy. This automated process can be used to detect small features and digitize other high-resolution drawings.

">

Geological Map Feature Extraction using Object Detection Techniques - A Comparative Analysis

P. A. N. Dilhan*, R. Siyambalapitiya**
*-** Department of Statistics and Computer Science, Faculty of Science, University of Peradeniya, Peradeniya, Sri Lanka.
Periodicity:April - June'2022
DOI : https://doi.org/10.26634/jip.9.2.18916

Abstract

Conducting a geological field survey at the initial stage is an important step in geo-oriented projects and construction. Therefore, better and more accurate solutions are only possible with field analysis and proper modeling. The geological modeling process takes a long time, especially depending on the area of interest. It is inefficient to digitize 2D geological maps with traditional software that uses manual user interaction. This paper proposes a state-of-the-art feature detection methodology for detecting geological features on high-resolution maps. With the development of efficient deep learning algorithms and the improvement of hardware systems, the accuracy of detecting specific objects in digital images, such as human facial features, has reached more than 90%. Current object detection models based on convolutional neural networks cannot be directly applied to high-resolution geological maps due to the input image size limitations of conventional object detection solutions, mostly limited by hardware resources. This paper proposes a sliding window method for character detection of geological features. Detection models are trained using transfer learning with You Look Only Once-v3 (YOLO-v3), Single Shot Multi-Box Detector (SSD), Faster-Region-based Convolutional Neural Network (Faster-RCNN), and Single Shot Multi-Box Detector_RetinaNet (SSD_RetinaNet). All models provide competitive success rates with an average precision (AP) of 0.96 on YOLOv3, 0.88 AP on EfficientNet, 0.92 AP on Faster- RCNN, and 0.97 AP on SSD_RetinaNet. YOLOv3 outperformed the best detection over SSD according to F1 recall and score. Since the input size of detection models is limited, a sliding window algorithm is used to separate high-resolution map images. The final detected strike features are provided as a digital dataset that can be used for further manipulations. Thus, Convolutional Neural Network (CNN) based object detection along with a sliding window protocol can be applied to manual map digitization processes to provide instantaneous digitized data with higher accuracy. This automated process can be used to detect small features and digitize other high-resolution drawings.

Keywords

You Look Only Once-v3, Faster-RCNN, Single Shot Multi-Box Detector, Convolutional Neural Network.

How to Cite this Article?

Dilhan, P. A. N., and Siyambalapitiya, R. (2022). Geological Map Feature Extraction using Object Detection Techniques - A Comparative Analysis. i-manager’s Journal on Image Processing, 9(2), 28-36. https://doi.org/10.26634/jip.9.2.18916

References

[1]. Alganci, U., Soydas, M., & Sertel, E. (2020). Comparative research on deep learning approaches for airplane detection from very high-resolution satellite images. Remote Sensing, 12(3), 458. https://doi.org/10.3390/rs12030458
[2]. Comaschi, F., Stuijk, S., Basten, T., & Corporaal, H. (2013). RASW: a run-time adaptive sliding window to improve viola-jones object detection. In 2013 Seventh International Conference on Distributed Smart Cameras (ICDSC), 1-6. https://doi.org/10.1109/ICDSC.2013.6778224
[3]. Farahbakhsh, E., Chandra, R., Olierook, H. K., Scalzo, R., Clark, C., Reddy, S. M., & Müller, R. D. (2020). Computer vision-based framework for extracting tectonic lineaments from optical remote sensing data. International Journal of Remote Sensing, 41(5), 1760-1787. https://doi.org/10.1080/01431161.2019.1674462
[4]. Geological Survey of Japan. (n.d.). Geological Maps Download. Retrieved from https://gbank.gsj.jp/datastore/consent.php?lang=en
[5]. Github. (n.d). Tensor Flow 2 Detection Model Zoo. Retrieved from https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md
[6]. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770-778.
[7]. Helvaci, C., & Bayram, B. (2004). Semi automatic digitizing of contours from 1: 25000 scaled maps. Proceedings of the International Archives of the Photogrammetr y, Remote Sensing and Spatial Information Sciences (ISPRS), 34, 12-23.
[8]. Li, K., Wan, G., Cheng, G., Meng, L., & Han, J. (2020). Object detection in optical remote sensing images: A survey and a new benchmark. ISPRS Journal of Photogrammetry and Remote Sensing, 159, 296-307. https://doi.org/10.1016/j.isprsjprs.2019.11.023
[9]. Lin, B., Zhou, L., Lv, G., & Zhu, A. X. (2017a). 3D geological modelling based on 2D geological map. Annals of GIS, 23(2), 117-129. https://doi.org/10.1080/19475683.2017.1304450
[10]. Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017b). Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision, 2980-2988.
[11]. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016). SSD: Single shot multibox detector. In European Conference on Computer Vision, 9905, 21-37. https://doi.org/10.1007/978-3-319-46448-0_2
[12]. Pradhan, R., Kumar, S., Agarwal, R., Pradhan, M. P., & Ghose, M. K. (2010). Contour line tracing algorithm for digital topographic maps. International Journal of Image Processing (IJIP), 4(2), 156-163.
[13]. Redmon, J., & Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv preprint arXiv: 1804.02767. https://doi.org/10.48550/arXiv.1804.02767
[14]. Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster RCNN: Towards real-time object detection with region proposal networks. Advances in Neural Information Processing Systems, 28.
[15]. Tan, L., Huangfu, T., Wu, L., & Chen, W. (2021). Comparison of RetinaNet, SSD, and YOLO v3 for real-time pill identification. BMC Medical Informatics and Decision Making, 21(1), 1-11. https://doi.org/10.1186/s12911-021-01691-8
[16]. Tan, M., & Le, Q. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. In International Conference on Machine Learning, 6105-6114.
If you have access to this article please login to view the article or kindly login to purchase the article

Purchase Instant Access

Single Article

North Americas,UK,
Middle East,Europe
India Rest of world
USD EUR INR USD-ROW
Pdf 35 35 200 20
Online 35 35 200 15
Pdf & Online 35 35 400 25

Options for accessing this content:
  • If you would like institutional access to this content, please recommend the title to your librarian.
    Library Recommendation Form
  • If you already have i-manager's user account: Login above and proceed to purchase the article.
  • New Users: Please register, then proceed to purchase the article.