Designing an Image Classification Model on Emergency Incident Images using a Convolutional Neural Network for iRESPOND

Designing an Image Classification Model on Emergency Incident Images using a Convolutional Neural Network for iRESPOND
  IJETT-book-cover           
  
© 2024 by IJETT Journal
Volume-72 Issue-3
Year of Publication : 2024
Author : Freddie Prianes, Ichelle Baluis, Joseph Jessie Oñate, Challiz Omorog, Thelma Palaoag, Nancy Flores
DOI : 10.14445/22315381/IJETT-V72I3P128

How to Cite?

Freddie Prianes, Ichelle Baluis, Joseph Jessie Oñate, Challiz Omorog, Thelma Palaoag, Nancy Flores, "Designing an Image Classification Model on Emergency Incident Images using a Convolutional Neural Network for iRESPOND," International Journal of Engineering Trends and Technology, vol. 72, no. 3, pp. 318-330, 2024. Crossref, https://doi.org/10.14445/22315381/IJETT-V72I3P128

Abstract
In the face of increasing natural disasters and emergencies, there is a growing need for effective geospatial information systems to process and classify emergency reports in real time. This work presents a new Convolutional Neural Network (CNN) model that is intended to classify emergency images taken and delivered to the iRESPOND system. Through the utilization of training phases and various tools, frameworks, and techniques, the authors effectively used deep learning to develop the CNN model. This model improves disaster response and mitigation by enabling the iRESPOND system to categorize emergency incidents rapidly. The results showcase the model’s commendable performance, achieving a high accuracy of 95.02% on the test set. A comprehensive evaluation, including precision, recall, and F1-score metrics for individual classes, illuminates the model’s strengths and areas for improvement. Noteworthy is the model’s proficiency in classes such as ‘flood’, ‘infrastructure_damage’, ‘no_damage_buildings_street’, ‘no_damage_ water_related’, and ‘no_damage_wildlife_forest’, reflecting robust predictive capabilities in specific emergency scenarios. The interpretability of the CNN model is augmented through visualization techniques like LIME, Grad-CAM, and Grad-CAM++. Also, a visualization report featuring the original image alongside interpretability overlays provides information on the characteristics and areas of the original images that influence the model’s decisions. In conclusion, the model demonstrates efficacy in rapidly categorizing emergency incidents, providing a valuable tool for the response team. The recommendations for future work underscore the continuous refinement required for optimal performance, including addressing class imbalances, fine-tuning hyperparameters, exploring ensemble models, and expanding the diverse image datasets.

Keywords
Emergency response, Machine learning, Resnet50, Tensorflow, Keras.

References
[1] Mary Ann E. Ignaco, “Mobile Application for Incident Reporting,” International Journal on Informatics Visualization, vol. 5, no. 4, pp. 1-7, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Barry Sheehan et al., “On the Benefits of Insurance and Disaster Risk Management Integration for Improved Climate-Related Natural Catastrophe Resilience,” Environment Systems and Decisions, vol. 43, pp. 639-648, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Global Assessment Report on Disaster Risk Reduction, GAR Special Report on Drought 2021, United Nations Office for Disaster Risk Reduction. [Online]. Available: https://www.undrr.org/publication/gar-special-report-drought-2021
[4] Trends in Disasters - Disasters in 2020 and 2021, World Disaster Report, pp. 210-243, 2022. [Online]. Available: https://www.ifrc.org/sites/default/files/2023-01/20230130_2022_WDR_DataAnnex.pdf
[5] Center for Fire Statistics of CTIF, World Fire Statistics, pp. 1-144, 2023. [Online]. Available: https://www.ctif.org/sites/default/files/2023-06/CTIF_Report28-ESG.pdf
[6] Yifan Xu et al., “Global, Regional, and National Burden of Road Injuries from 1990 to 2019,” International Journal of Environment Research and Public Health, vol. 19, no. 24, pp. 1-20, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Ralf Bill et al., “Geospatial Information Research: State of the Art, Case Studies and Future Perspectives,” PFG-Journal of Photogrammetry, Remote Sensing and Geoinformation Science, vol. 90, pp. 349-389, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Robertas Damaševičius, Nebojsa Bacanin, and Sanjay Misra, “From Sensors to Safety: Internet of Emergency Services (IoES) for Emergency Response and Disaster Management,” Journal of Sensor and Actuator Networks, vol. 12, no. 3, pp. 1-45, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Yann LeCun, Yoshua Bengio, and Geoffrey Hinton, “Deep Learning,” Nature, vol. 521, pp. 436-444, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Saman Ghaffarian, Firouzeh Rosa Taghikhah, and Holger R. Maier, “Explainable Artificial Intelligence in Disaster Risk Management: Achievements and Prospective Futures,” International Journal of Disaster Risk Reduction, vol. 98, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Saman Ghaffarian et al., “Post-Disaster Building Database Updating Using Automated Deep Learning: An Integration of Pre-Disaster OpenStreetMap and Multi-Temporal Satellite Data,” Remote Sensing, vol. 11, no. 20, pp. 1-20, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Yongqing Tang et al., “ForestResNet: A Deep Learning Algorithm for Forest Image Classification,” Journal of Physics, vol. 2024, no. 1, pp. 1-6, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Archit Rathod et al., “Leveraging CNNs and Ensemble Learning for Automated Disaster Image Classification,” arXiv, vol. 1, pp. 1-13, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Jash Shah et al., “Proposed Methodology for Disaster Classification Using Computer Vision and Federated Learning,” Research Square, pp. 1-9, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Laith Alzubaidi et al., “Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Directions,” Journal of Big Data, vol. 8, no. 53, pp. 1-74, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Amna Asif et al., “Automatic Analysis of Social Media Images to Identify Disaster Type and Infer Appropriate Emergency Response,” Journal of Big Data, vol. 8, no. 83, pp. 1-28, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Shannon Daly, and James A. Thom, “Mining and Classifying Image Posts on Social Media to Analyse Fires,” ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management, pp. 1-14, 2016.
[Google Scholar] [Publisher Link]
[18] Ayan Mukhopadhyay et al., “A Review of Incident Prediction, Resource Allocation, and Dispatch Models for Emergency Management,” Accident Analysis & Prevention, vol. 165, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[19] J. Kallas, and R. Napolitano, “Automated Large-Scale Damage Detection on Historic Buildings in Post-Disaster Areas Using Image Segmentation,” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 48, pp. 797- 804, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Neha Sharma, Vibhor Jain, and Anju Mishra, “An Analysis of Convolutional Neural Networks for Image Classification,” Procedia Computer Science, vol. 132, pp. 377-384, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Siti Maisarah Zainorzuli et al., “Comparison Study on Convolution Neural Network (CNN) Techniques for Image Classification,” Journal of Electrical and Electronic Systems Research, vol. 20, pp. 11-17, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[22] L. Sushma, and K.P. Lakshmi, “An Analysis of Convolution Neural Network for Image Classification using Different Models,” International Journal of Engineering Research & Technology (IJERT), vol. 9, no. 10, pp. 629-637, 2020.
[Google Scholar] [Publisher Link]
[23] Sendai Framework for Disaster Risk Reduction 2015-2030, United Nation Office for Disaster Risk Reduction, 2019. [Online]. Available: https://www.undrr.org/publication/sendai-framework-disaster-risk-reduction-2015-2030
[24] Martín Abadi et al., “TensorFlow: A System for Large-Scale Machine Learning,” OSDI'16: Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation, pp. 265-283, 2016.
[Google Scholar] [Publisher Link]
[25] Simple, Flexible, Powerful, Keras, 2023. [Online]. Available: https://keras.io.
[26] Kaiming He et al., “Deep Residual Learning for Image Recognition,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770-778, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Diederik P. Kingma, and Jimmy Ba, “Adam: A Method for Stochastic Optimization,” 3 rd International Conference on Learning Representation, San Diego, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin, “Why Should I Trust You?”: Explaining the Predictions of Any Classifier,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16), pp. 1135-1144, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Ramprasaath R. Selvaraju et al., “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 618-626, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[30] Aditya Chattopadhay et al., “Grad-CAM++: Generalized Gradient-Based Visual Explanations for Deep Convolutional Networks,” 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Tahoe, NV, USA, pp. 839-847, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[31] Jia Deng et al., “ImageNet: A Large-Scale Hierarchical Image Database,” 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, pp. 248-255, 2009.
[CrossRef] [Google Scholar] [Publisher Link]
[32] James Bergstra, and Yoshua Bengio, “Random Search for Hyper-Parameter Optimization,” Journal of Machine Learning Research, vol. 13, no. 10, pp. 281-305, 2012.
[Google Scholar] [Publisher Link]
[33] Holger R. Maier et al., “On How Data are Partitioned in Model Development and Evaluation: Confronting the Elephant in the Room to Enhance Model Generalization,” Environmental Modelling & Software, vol. 167, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[34] John D. Hunter, “Matplotlib: A 2D Graphics Environment,” Computing in Science & Engineering, vol. 9, no. 3, pp. 90-95, 2007.
[CrossRef] [Google Scholar] [Publisher Link]
[35] David M.W. Powers, “Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness & Correlation,” International Journal of Machine Learning Technologies, vol. 2, no. 1, pp. 37-63, 2011.
[CrossRef] [Google Scholar] [Publisher Link]