The week's pick
Reseach Article
TerraGrid: Harnessing Deep Learning Models for Satellite Image Segmentation
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 186 - Number 49 |
| Year of Publication: 2024 |
| Authors: Nitin Pal, Soumya Ramkrishna, Harsh Patil, Nishant Choudhary, Rajashree Soman |
10.5120/ijca2024924147
|
Nitin Pal, Soumya Ramkrishna, Harsh Patil, Nishant Choudhary, Rajashree Soman . TerraGrid: Harnessing Deep Learning Models for Satellite Image Segmentation. International Journal of Computer Applications. 186, 49 ( Nov 2024), 14-21. DOI=10.5120/ijca2024924147
Abstract
Satellite imaging is a backbone of environmental monitoring and urban planning. State-of-the-art methodologies for image segmentation are extremely needed to process huge datasets in order to obtain substantial information out of it. This paper presents an extensive analysis on the current state of satellite image segmentation using deep learning with a special emphasis on some CNN architectures, namely InceptionResNetV2, InceptionResNetV2-UNet, Multi-UNet, VGG19, and VGG19-UNet. In this work, a number of pre-processing techniques were used namely advanced data augmentation and normalization. The set of experiments is done in the motivation to evaluate models in the best way based on a few performance metrics, such as accuracy, Dice coefficient, and validation loss among others, compared with the other models. The conducted experiments have shown that the InceptionResNetV2-UNet and VGG19-UNet provide higher segmentation accuracy compared to other models by approving a test accuracy of 89.90% and 89.41% with the dice coefficient of 85.5% and 84.9%, respectively. Further from this, proposed work introduce a Gradio-based web application for the end users to predict segmented images interactively, demonstrating real-world use cases for employed models. This bridges the gap between advanced machine learning research and satellite image evaluation, providing valuable insights for future work in this field.
References
- Vance, T. C., Huang, T., and Butler, K. A. 2024. Big data in Earth science: Emerging practice and promise. Science.
- Garea, S. A. and Das, S. 2024. Image Segmentation Methods: Overview, Challenges, and Future Directions. 2024 Seventh International Women in Data Science Conference at Prince Sultan University (WiDS PSU).
- Tong, X., Xia, G., Lu, Q., Shen, H., Li, S. You S., and Zhang, L. 2020. Land-cover classification with high-resolution remote sensing images using transferable deep models. Remote Sensing of Environment.
- Toennies, K. D. 2024. Image Classification: A Computer Vision Task. In: An Introduction to Image Classification. Springer.
- Pearce, W., Özkula, S. M., Greene, A. K., Teeling, L., Bansard, J. S., Omena J. J., and Rabello, E. T. 2018. Visual cross-platform analysis: digital methods to research social media images. Information. Communication & Society.
- Sheth, V., Tripathi, U., and Sharma, A. 2022. A Comparative Analysis of Machine Learning Algorithms for Classification Purpose. Procedia Computer Science.
- Aghayari, S., Hadavand, A., Mohamadnezhad Niazi, S., and Omidalizarandi, M. 2023. Building Detection from Aerial Imagery using Inception Resnet UNET and UNET Models. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences.
- Peng, C., Liu, Y., Yuan X., and Chen, Q., 2022. Research of image recognition method based on enhanced inception-ResNet-V2. Multimedia Tools Application.
- Alfarhan, M., Deriche, M., and Maalej, A. 2020. Robust Concurrent Detection of Salt Domes and Faults in Seismic Surveys Using an Improved UNet Architecture. IEEE Access.
- Haneen, A., and Ahmad. M. B. 2022. Deep Learning-Based Frameworks for Semantic Segmentation of Road Scenes. Electronics.
- Busra, E. S., Guzel, M. S., Bostanci, G. E., Ekinci, F., Asuroglu, T., and Acici, K. 2023. Deep-Learning-Based Approaches for Semantic Segmentation of Natural Scene Images: A Review. Electronics.
- Bouhsissin, S., Sael N., and Benabbou, F. 2021. Enhanced VGG19 Model for Accident Detection and Classification from Video. 2021 International Conference on Digital Age & Technological Advances for Sustainable Development (ICDATA).
- Maghdid H. S., Asaad A. T., Ghafoor K. Z., Sadiq A. S., and Khan M. K. 2020. Diagnosing COVID-19 pneumonia from X-ray and CT images using deep learning and transfer learning algorithms. CoRR ArXiv.
- Ghoshal, B., and Tucker, A. 2020. Estimating uncertainty and interpretability in deep learning for coronavirus (COVID-19) detection. ArXiv.
- Shankar, K., Zhang, Y., Liu, Y., Wu, L., Chen, C. H. 2020. Hyperparameter tuning deep learning for diabetic retinopathy fundus image classification. IEEE Access.
- Ji, S., Wei, S., and Lu, M. 2019. Fully Convolutional Networks for Multisource Building Extraction from an Open Aerial and Satellite Imagery Data Set. In IEEE Transactions on Geoscience and Remote Sensing.
- Jiwani, A., Ganguly, S., Ding, C., Zhou, N., and Chan, D. M. 2021. A Semantic Segmentation Network for Urban-Scale Building Footprint Extraction Using RGB Satellite Imagery. ArXiv.
- Maggiori, E., Tarabalka, Y., Charpiat, G., and Alliez, P. 2017. Can semantic labeling methods generalize to any city? the inria aerial image labeling benchmark. 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).
- Chen, Q., Wang, L., Wu, Y., Wu, G., Guo, Z., and Waslander, S. L. 2019. Aerial imagery for roof segmentation: A large-scale dataset towards automatic mapping of buildings. ISPRS Journal of Photogrammetry and Remote Sensing.
- Cai, Y., He, H., Yang, K., Fatholahi, S. N., Ma, L., Xu, L., and Li, J. 2021. A Comparative Study of Deep Learning Approaches to Rooftop Detection in Aerial Images. Canadian Journal of Remote Sensing.
- Glinka, S., Owerko, T., and Tomaszkiewicz, K. 2022. Using Open Vector-Based Spatial Data to Create Semantic Datasets for Building Segmentation for Raster Data. Remote Sensing.
- Bakirman, T., Komurcu, I., and Sertel, E. 2022. Comparative analysis of deep learning based building extraction methods with the new VHR Istanbul dataset. Expert Systems with Applications.
- Amirgan, B., and Erener, A. 2024. Semantic segmentation of satellite images with different building types using deep learning methods. Remote Sensing Applications: Society and Environment.
- Alexander, B., Vladimir, I., Eugene, K., Alex, P., Mikhail, D., and Alexandr, K. 2020. Albumentations: Fast and Flexible Image Augmentations, Information.
- Saponara, S., and Elhanashi, A. 2021. Impact of Image Resizing on Deep Learning Detectors for Training Time and Model Performance. In: Saponara, S., De Gloria, A. (eds) Applications in Electronics Pervading Industry, Environment and Society. ApplePies 2021, Springer.
- Siciarz, P., and McCurdy, B. 2022. U-net architecture with embedded Inception-ResNet-v2 image encoding modules for automatic segmentation of organs-at-risk in head and neck cancer radiation therapy based on computed tomography scans. Physics in Medicine & Biology.
- Anaya-Isaza, A., Mera-Jiménez, L., Cabrera-Chavarro, J. M., Guachi-Guachi, L., Peluffo-Ordóñez, D., and Rios-Patiño, J. I. 2021. Comparison of Current Deep Convolutional Neural Networks for the Segmentation of Breast Masses in Mammograms. In IEEE Access.
- Zhang Y., and Guindon, B. 2016. Application of the Dice Coefficient to Accuracy Assessment of Object-Based Image Classification. Canadian Journal of Remote Sensing.
- Szegedy, C., Wei, L., Yangqing, J., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., and Rabinovich, A. 2015. Going deeper with convolutions. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
- Szegedy, C., Ioffe, S., Vanhoucke, V., and Alemi, A. A. 2017. Inception-v4, inception-ResNet and the impact of residual connections on learning. In Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI'17), AAAI Press.
- Ronneberger, O., Fischer, P., and Brox, T. 2015. UNet: Convolutional networks for biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention. Springer.
Index Terms
Keywords