Terrain change monitoring from multi-temporal aerospace images using advanced training neural networks

Aerophototopography (Aerial Surveying And Photogrammetry)
UDC: 
528.7
DOI: 
10.22389/0016-7126-2025-1019-5-39-46
1 Ivanov V.G.
2 Kuzina E.I.
3 Gomanov D.E.
4 Grishin P.S.
5 Cheglov A.V.
6 Kuzin P.I.
Year: 
2025
№: 
5
1019
Pages: 
39-46

Military Academy of Communication named after Marshal of the Soviet Union S. M. Budyonny

1, 
2, 

Military Topographic Directorate of the General Staff of the Armed Forces of the Russian Federation

3, 

"18 Central Research Institute" Ministry of Defence of the Russian Federation

4, 
5, 

St. Petersburg State Forest Technical University named after S. M. Kirov

6, 
Abstract:
One of the key elements of the digital map making process is formation and updating of objects by detecting terrain changes from multi-temporal highly detailed aerospace images. At the same time, at present the main predominant method of change detecting the said images is still their direct visual analysis, implemented by the deciphering operator of the ground complex of view information processing, which, in general, reduces the efficiency of digital maps preparation. Application of progressive technologies of automatic change detection will allow ensuring a better quality of digital maps, including those according to heterogeneous remote sensing data (radar imagery, laser scanning). Methods based on artificial intelligence technologies, in particular advanced training of neural networks, show significant ‘stability’, which is due to formation of the most informative signs of significant changes in the process of neural network training. The proposed developed ResNet-50-CDB neural network and algorithm for reversible processing of input images will provide lower probability of errors of the first and second type, which gives both higher Jaccard index and accuracy of boundary localisation
References: 
1.   Guk A.P., Evstratova L.G., Khlebnikov E.P., Altyntsev M.A., Arbuzov S.A., Gordienko A.S., Guk A.A. (2013) Automated interpretation of space images. Detection of changes in the status of territories and objects on multispectral satellite images obtained at different dates. Geodezia i Kartografia, (8), pp. 39-47.
2.   Zharova N.E., Bekenov A.V., Chibunichev Aleksandr (2017) Automatic generation of a digital terrain model from «Resurs-P» stereo «fortuitous» image pair. Geodezia i Kartografia, 78(10), pp. 50-57 . (In Russian). DOI: 10.22389/0016-7126-2017-928-10-50-57.
3.   Kobzev A. A., Chibunichev A. G. Issledovanie vozmozhnosti sovmestnoi fotogrammetricheskoi obrabotki raznovremennykh aerosnimkov. Izvestia vuzov. Geodesy and Aerophotosurveying, 2021, no. 3, pp. 292–301.
4.   Permyakov R.V. (2021) Photogrammetric processing and application of multi-temporal satellite imagery stereopairs. Geodezia i Kartografia, 82(8), pp. 36-44. (In Russian). DOI: 10.22389/0016-7126-2021-974-8-36-44.
5.   Chen H., Wu C., Du B., Zhang L. (2019) Deep Siamese Multi-Scale Convolutional Network for Change Detection in Multi-temporal VHR Images. IEEE Transactions on Image Processing, DOI: 10.48550/arXiv.1906.11479.
6.   Daudt R. C., Le Saux B., Boulch A. (2018) Fully Convolutional Siamese Networks for Change Detection. Proceedings of the 25th IEEE International Conference on Image Processing (Athens, Greece, 7–10 October 2018). no. 36, pp. 4063–4067. DOI: 10.48550/arXiv.1810.08462.
7.   Daudt R. C., Le Saux B., Boulch A., Gousseau Y. (2018) Urban change detection for multispectral earth observation using convolutional neural networks. International Geoscience and Remote Sensing Symposium (Valencia, Spain, 22–27 July 2018). pp. 2115–2118. DOI: 10.1109/IGARSS.2018.8518015.
8.   Ghosh S., Mishra N. S., Ghosh A. (2009) Unsupervised Change Detection of Remotely Sensed Images using Fuzzy Clustering. Seventh International Conference on Advances in Pattern Recognition (Kolkata, India). pp. 385–388. DOI: 10.1109/ICAPR.2009.82.
9.   He K., Zhang X., Ren S., Sun J. (2015) Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (Boston, MA, USA, 7–12 June). pp. 770–778.
10.   Hu B., Lu Z., Li H., Chen Q. (2014) Convolutional Neural Network Architectures for Matching Natural Language Sentences. Advances in Neural Information Processing Systems, no. 27, DOI: 10.48550/arXiv.1503.03244.
11.   Hussain M., Chen D., Cheng A., Wei H., Stanley D. (2013) Change detection from remotely sensed images: From pixel-based to object-based approaches. ISPRS Journal of Photogrammetry and Remote Sensing, no. 80, pp. 91–106. DOI: 10.1016/j.isprsjprs.2013.03.006.
12.   Ilsever M., Ünsalan C. (2012) Pixel-Based Change Detection Methods. Springer Briefs in Computer Science. Springer, London, pp. 7–21. DOI: 10.1007/978-1-4471-4255-3_2.
13.   Khurman M., Saxena V. (2015) Soft Computing Techniques for Change Detection in remotely sensed images. International Journal of Computational Science, Volume 12, no. 2, pp. 245–253.
14.   Patra S., Ghosh A., Ghosh S. (2008) Change Detection of Remote Sensing Images with Semi-Supervised Multilayer Perceptron. Fundamenta Informaticae, no. 84 (3-4), pp. 429–442. DOI: 10.3233/FUN-2008-843-409.
15.   Simonyan K., Zisserman A. (2015) Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv: 1409.1556.
16.   Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Erhan D., Vanhoucke V., Rabinovich A. (2014) Going deeper with convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (Columbus, OH, USA, 24–27 June 2014). DOI: 10.1109/CVPR.2015.7298594.
17.   Tan K., Wu F., Du Q., Du P., Chen Y. (2019) A Parallel Gaussian–Bernoulli Restricted Boltzmann Machine for Mining Area Classification with Hyperspectral Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote, Volume 12, no. 2, pp. 627–636. DOI: 10.1109/JSTARS.2019.2892975.
18.   Volpi M., Tuia D., Bovolo F., Kanevski M., Bruzzone L. (2013) Supervised change detection in VHR images using contextual information and support vector machines. International Journal of Applied Earth Observation and Geoinformation, no. 20, pp. 77–85. DOI: 10.1016/j.jag.2011.10.013.
19.   Wang X., Tan K., Du Q., Chen Y., Du P. (2019) Caps-Triple GAN: GAN-Assisted Caps Net for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, Volume 57, no. 9, pp. 7232–7245. DOI: 10.1109/TGRS.2019.2912468.
20.   Wang M., Tan K., Jia X., Wang X., Chen Y. (2020) A Deep Siamese Network with Hybrid Convolutional Feature Extraction Module for Change Detection Based on Multi-Sensor Remote Sensing Images. Remote Sensing, no. 12 (2), DOI: 10.3390/rs12020205.
Citation:
Ivanov V.G., 
Kuzina E.I., 
Gomanov D.E., 
Grishin P.S., 
Cheglov A.V., 
Kuzin P.I., 
(2025) Terrain change monitoring from multi-temporal aerospace images using advanced training neural networks. Geodesy and cartography = Geodeziya i Kartografiya, 86(5), pp. 39-46. (In Russian). DOI: 10.22389/0016-7126-2025-1019-5-39-46
Publication History
Received: 14.02.2025
Accepted: 12.05.2025
Published: 20.06.2025

Content

2025 May DOI:
10.22389/0016-7126-2025-1019-5