Localizing and Quantifying Damage in Social Media Images

Localizing and Quantifying Damage in Social Media Images. Li, X., Zhang, H., Caragea, D., & Imran, M. June, 2018.

Traditional post-disaster assessment of damage heavily relies on expensive GIS data, especially remote sensing image data. In recent years, social media has become a rich source of disaster information that may be useful in assessing damage at a lower cost. Such information includes text (e.g., tweets) or images posted by eyewitnesses of a disaster. Most of the existing research explores the use of text in identifying situational awareness information useful for disaster response teams. The use of social media images to assess disaster damage is limited. In this paper, we propose a novel approach, based on convolutional neural networks and class activation maps, to locate damage in a disaster image and to quantify the degree of the damage. Our proposed approach enables the use of social network images for post-disaster damage assessment and provides an inexpensive and feasible alternative to the more expensive GIS approach.

@article{li_localizing_2018,
	title = {Localizing and {Quantifying} {Damage} in {Social} {Media} {Images}},
	url = {https://arxiv.org/abs/1806.07378v1},
	abstract = {Traditional post-disaster assessment of damage heavily relies on expensive
GIS data, especially remote sensing image data. In recent years, social media
has become a rich source of disaster information that may be useful in
assessing damage at a lower cost. Such information includes text (e.g., tweets)
or images posted by eyewitnesses of a disaster. Most of the existing research
explores the use of text in identifying situational awareness information
useful for disaster response teams. The use of social media images to assess
disaster damage is limited. In this paper, we propose a novel approach, based
on convolutional neural networks and class activation maps, to locate damage in
a disaster image and to quantify the degree of the damage. Our proposed
approach enables the use of social network images for post-disaster damage
assessment and provides an inexpensive and feasible alternative to the more
expensive GIS approach.},
	language = {en},
	urldate = {2019-03-11},
	author = {Li, Xukun and Zhang, Huaiyu and Caragea, Doina and Imran, Muhammad},
	month = jun,
	year = {2018},
}

Downloads: 0

{"_id":"2e8mj5zRXNYHsDwjX","bibbaseid":"li-zhang-caragea-imran-localizingandquantifyingdamageinsocialmediaimages-2018","author_short":["Li, X.","Zhang, H.","Caragea, D.","Imran, M."],"bibdata":{"bibtype":"article","type":"article","title":"Localizing and Quantifying Damage in Social Media Images","url":"https://arxiv.org/abs/1806.07378v1","abstract":"Traditional post-disaster assessment of damage heavily relies on expensive GIS data, especially remote sensing image data. In recent years, social media has become a rich source of disaster information that may be useful in assessing damage at a lower cost. Such information includes text (e.g., tweets) or images posted by eyewitnesses of a disaster. Most of the existing research explores the use of text in identifying situational awareness information useful for disaster response teams. The use of social media images to assess disaster damage is limited. In this paper, we propose a novel approach, based on convolutional neural networks and class activation maps, to locate damage in a disaster image and to quantify the degree of the damage. Our proposed approach enables the use of social network images for post-disaster damage assessment and provides an inexpensive and feasible alternative to the more expensive GIS approach.","language":"en","urldate":"2019-03-11","author":[{"propositions":[],"lastnames":["Li"],"firstnames":["Xukun"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Huaiyu"],"suffixes":[]},{"propositions":[],"lastnames":["Caragea"],"firstnames":["Doina"],"suffixes":[]},{"propositions":[],"lastnames":["Imran"],"firstnames":["Muhammad"],"suffixes":[]}],"month":"June","year":"2018","bibtex":"@article{li_localizing_2018,\n\ttitle = {Localizing and {Quantifying} {Damage} in {Social} {Media} {Images}},\n\turl = {https://arxiv.org/abs/1806.07378v1},\n\tabstract = {Traditional post-disaster assessment of damage heavily relies on expensive\nGIS data, especially remote sensing image data. In recent years, social media\nhas become a rich source of disaster information that may be useful in\nassessing damage at a lower cost. Such information includes text (e.g., tweets)\nor images posted by eyewitnesses of a disaster. Most of the existing research\nexplores the use of text in identifying situational awareness information\nuseful for disaster response teams. The use of social media images to assess\ndisaster damage is limited. In this paper, we propose a novel approach, based\non convolutional neural networks and class activation maps, to locate damage in\na disaster image and to quantify the degree of the damage. Our proposed\napproach enables the use of social network images for post-disaster damage\nassessment and provides an inexpensive and feasible alternative to the more\nexpensive GIS approach.},\n\tlanguage = {en},\n\turldate = {2019-03-11},\n\tauthor = {Li, Xukun and Zhang, Huaiyu and Caragea, Doina and Imran, Muhammad},\n\tmonth = jun,\n\tyear = {2018},\n}\n\n","author_short":["Li, X.","Zhang, H.","Caragea, D.","Imran, M."],"key":"li_localizing_2018","id":"li_localizing_2018","bibbaseid":"li-zhang-caragea-imran-localizingandquantifyingdamageinsocialmediaimages-2018","role":"author","urls":{"Paper":"https://arxiv.org/abs/1806.07378v1"},"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/wybert","dataSources":["TJkbwzD8s2wCxBy6Y"],"keywords":[],"search_terms":["localizing","quantifying","damage","social","media","images","li","zhang","caragea","imran"],"title":"Localizing and Quantifying Damage in Social Media Images","year":2018}