{"id":581342,"date":"2019-08-09T12:10:58","date_gmt":"2019-08-09T19:10:58","guid":{"rendered":"https:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&#038;p=581342"},"modified":"2019-08-11T12:48:59","modified_gmt":"2019-08-11T19:48:59","slug":"damage-assessment-using-deep-learning-in-arcgis","status":"publish","type":"blog","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis","title":{"rendered":"Damage assessment using deep learning in ArcGIS"},"author":8452,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","format":"standard","meta":{"_acf_changed":false,"_searchwp_excluded":""},"categories":[22931],"tags":[42181,186132,445012,115262,35661],"industry":[],"product":[36841,36561],"class_list":["post-581342","blog","type-blog","status-publish","format-standard","hentry","category-imagery","tag-arcgis-pro","tag-deep-learning","tag-image-analyst-extension","tag-imagery","tag-machine-learning","product-api-python","product-arcgis-pro"],"acf":{"short_description":"Here's a look at how the damage assessment deep learning demo was done by USAA at the UC2019 plenary.","flexible_content":[{"acf_fc_layout":"content","content":"<p>In the aftermath of a natural disaster, response and recovery efforts can be drastically slowed down by manual data collection. Traditionally, insurance assessors and government officials have to rely on human interpretation of imagery and site visits to assess damage and loss. But depending on the scope of a disaster, this necessary process could delay relief to disaster victims.<\/p>\n<aside class=\"newsr-aside newsr-aside--end column-10 tablet-column-12 phone-column-6\">\n<div class=\"newsr-aside_custom-content\">\n<p><strong>Article Snapshot:<\/strong> At this year\u2019s Esri User Conference plenary session, the United Services Automobile Association (USAA) <a href=\"https:\/\/www.youtube.com\/watch?v=0OWKZOrKfv8&amp;list=PLaPDDLTCmy4bC2dGacC3ZH8YnCl4x7kuB&amp;index=15\">demonstrated the use of deep learning<\/a> capabilities in ArcGIS to perform automated damage assessment of homes after the devastating\u00a0<a href=\"https:\/\/en.wikipedia.org\/wiki\/Woolsey_Fire\">Woolsey fire<\/a>. This work was a collaborative prototype between Esri and USAA to show the art of the possible in doing this type of damage assessment using the ArcGIS platform.<\/p>\n<\/div>\n<\/aside>\n<p>The Woolsey Fire burned for 15 days, burning almost 97,000 acres, and damaging or destroying thousands of structures. Deep learning within ArcGIS was used to quickly identify damaged structures within the fire perimeter, fast tracking the time for impacted residents and businesses to have\u00a0their adjuster process the insurance claims.<\/p>\n<p>The process included capturing training samples, training the deep learning model, running inferencing tools and detecting damaged homes &#8211; all done within the ArcGIS platform. In this blog, we\u2019ll walk through each step in the process.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":583432,"id":583432,"title":"End-to-end deep learning in ArcGIS","filename":"Workflow2.png","filesize":173012,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/workflow2","alt":"","author":"8452","description":"","caption":"","name":"workflow2","status":"inherit","uploaded_to":581342,"date":"2019-08-09 19:05:40","modified":"2019-08-09 19:06:09","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1270,"height":201,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","medium-width":464,"medium-height":73,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","medium_large-width":768,"medium_large-height":122,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","large-width":1270,"large-height":201,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","1536x1536-width":1270,"1536x1536-height":201,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","2048x2048-width":1270,"2048x2048-height":201,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2-826x131.png","card_image-width":826,"card_image-height":131,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/Workflow2.png","wide_image-width":1270,"wide_image-height":201}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<h1><strong>Step1: Managing the imagery<\/strong><\/h1>\n<p>Before the fires were extinguished, DataWing flew drones in the fire perimeter and captured high resolution imagery of impacted areas. The imagery totaled 40 GB in size and was managed using a mosaic dataset. The mosaic dataset is the primary image management model for ArcGIS to manage large volumes of imagery.<\/p>\n<h1><strong>Step2. Labelling and preparing training samples<\/strong><\/h1>\n<p>Prior to training a deep learning model, training samples must be created to represent areas of interest &#8211; in this case, the USAA was interested in damaged and undamaged buildings. The <a href=\"https:\/\/egis3.lacounty.gov\/dataportal\/2011\/04\/28\/countywide-building-outlines\/\">building footprint data<\/a> provided by LA County, was overlaid on the high resolution drone imagery in ArcGIS Pro, and several hundred homes were manually labelled as Damaged or Undamaged \u00a0(a new field called \u201cClassValue\u201d in the building footprint feature class was attributed with this information). These training features were used to export <em>training samples <\/em>using the\u00a0<a href=\"https:\/\/pro.arcgis.com\/en\/pro-app\/tool-reference\/image-analyst\/export-training-data-for-deep-learning.htm\">Export Training Data for Deep Learning<\/a>\u00a0tool in ArcGIS Pro, with the metadata output format set to \u2018Labeled Tiles\u2019.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":581482,"id":581482,"title":"Damaged and undamaged houses after a wildfire","filename":"chips-e1565155751524.png","filesize":192099,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/chips","alt":"Damaged and undamaged houses after a wildfire","author":"8452","description":"Damaged and undamaged houses after a wildfire","caption":"Damaged and undamaged houses after a wildfire","name":"chips","status":"inherit","uploaded_to":581342,"date":"2019-08-07 05:27:15","modified":"2020-07-08 08:41:09","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":400,"height":229,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","medium-width":400,"medium-height":229,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","medium_large-width":400,"medium_large-height":229,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","large-width":400,"large-height":229,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","1536x1536-width":400,"1536x1536-height":229,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","2048x2048-width":400,"2048x2048-height":229,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","card_image-width":400,"card_image-height":229,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/chips-e1565155751524.png","wide_image-width":400,"wide_image-height":229}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<h1><strong>Step 3: Training the deep learning model<\/strong><\/h1>\n<p><a href=\"http:\/\/bit.ly\/2YCFHQn\">ArcGIS Notebooks<\/a> was used for training purposes. ArcGIS Notebooks is pre-configured with the necessary deep learning libraries, so no extra setup was required. With a few lines of code, the training samples exported from ArcGIS Pro were augmented. Using the arcgis.learn module in the ArcGIS Python API, optimum training parameters for the damage assessment model were set, and the deep learning model was trained using a ResNet34 architecture to classify all buildings in the imagery as either damaged or undamaged.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":583322,"id":583322,"title":"The model converged around 99% accuracy","filename":"modelConverge2.jpg","filesize":56051,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/modelconverge2","alt":"","author":"8452","description":"The model converged around 99% accuracy","caption":"The model converged around 99% accuracy","name":"modelconverge2","status":"inherit","uploaded_to":581342,"date":"2019-08-09 16:07:51","modified":"2019-08-09 16:08:06","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":580,"height":783,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","medium-width":193,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","medium_large-width":580,"medium_large-height":783,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","large-width":580,"large-height":783,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","1536x1536-width":580,"1536x1536-height":783,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","2048x2048-width":580,"2048x2048-height":783,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2-344x465.jpg","card_image-width":344,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/modelConverge2.jpg","wide_image-width":580,"wide_image-height":783}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<p>Once complete, the ground truth labels were compared to the model classification results to get a quick qualitative idea on how well the model performed.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":581532,"id":581532,"title":"Model Predictions","filename":"ModelPredicitons.jpg","filesize":57885,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/modelpredicitons","alt":"Model Predictions","author":"8452","description":"Model Predictions","caption":"Model Predictions","name":"modelpredicitons","status":"inherit","uploaded_to":581342,"date":"2019-08-07 05:36:13","modified":"2019-08-07 05:36:38","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1131,"height":381,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","medium-width":464,"medium-height":156,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","medium_large-width":768,"medium_large-height":259,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","large-width":1131,"large-height":381,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","1536x1536-width":1131,"1536x1536-height":381,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","2048x2048-width":1131,"2048x2048-height":381,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons-826x278.jpg","card_image-width":826,"card_image-height":278,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/ModelPredicitons.jpg","wide_image-width":1131,"wide_image-height":381}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<p><strong>For complete details on the training process see <a href=\"https:\/\/medium.com\/geoai\/building-footprint-extraction-and-damage-classification-8a5458759332\">our post<\/a> on Medium <\/strong><\/p>\n<p>Finally, with the model.save() function, the model can be saved and used for inferencing purposes.<\/p>\n<p>&nbsp;<\/p>\n<h1><strong>Step 4: Running the inferencing tools<\/strong><\/h1>\n<p>Inferencing was performed using the ArcGIS API for Python. By running inferencing inside of ArcGIS Enterprise using the model.classify_features function in Notebooks, we can take the inferencing to scale.<\/p>\n<p>The result is a feature service that can be viewed in ArcGIS Pro. (Here\u2019s a link to the <a href=\"https:\/\/myago.maps.arcgis.com\/home\/webmap\/viewer.html?webmap=ac86e722bb3d4f3cb6ffc906b69605f4\">web map<\/a>).<\/p>\n<p>Over nine thousand buildings were automatically classified using deep learning capabilities within ArcGIS!<\/p>\n<p>The map below shows the damaged buildings marked in red, and the undamaged buildings in green. With 99% accuracy, the model is approaching the performance of a trained adjuster &#8211; what used to take us days or weeks, now we can do in a matter of hours.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":581662,"id":581662,"title":"Damage assessment using object classification. Damaged houses are shown in red and undamaged ones in blue.","filename":"showingaccuracyinpro.gif","filesize":2060931,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/showingaccuracyinpro","alt":"Inference results","author":"8452","description":"","caption":"Damage assessment using object classification. Damaged houses are shown in red and undamaged ones in blue.\n","name":"showingaccuracyinpro","status":"inherit","uploaded_to":581342,"date":"2019-08-07 17:01:15","modified":"2020-07-09 04:10:35","menu_order":0,"mime_type":"image\/gif","type":"image","subtype":"gif","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":600,"height":320,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro-213x200.gif","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","medium-width":464,"medium-height":247,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","medium_large-width":600,"medium_large-height":320,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","large-width":600,"large-height":320,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","1536x1536-width":600,"1536x1536-height":320,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","2048x2048-width":600,"2048x2048-height":320,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","card_image-width":600,"card_image-height":320,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/showingaccuracyinpro.gif","wide_image-width":600,"wide_image-height":320}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<h1><strong>Step 5: Deriving valuable insights<\/strong><\/h1>\n<p><strong>Business Analyst:<\/strong> Now that we had a better understanding of the impacted area, we wanted to understand who were the members impacted by the fires. When deploying mobile response units to disaster areas, it\u2019s important to know where the most at-risk populations are located, for example, the elderly or children. Using Infographics from ArcGIS Business Analyst, we extracted valuable characteristics and information about the impacted community and generated a report to help mobile units make decisions faster.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":581682,"id":581682,"title":"Get location intelligence with ArcGIS Business Analyst","filename":"businessAnalyst.png","filesize":117072,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/businessanalyst-2","alt":"Get location intelligence with ArcGIS Business Analyst","author":"8452","description":"Get location intelligence with ArcGIS Business Analyst","caption":"Get location intelligence with ArcGIS Business Analyst","name":"businessanalyst-2","status":"inherit","uploaded_to":581342,"date":"2019-08-07 17:05:03","modified":"2019-08-07 17:08:02","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1366,"height":768,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","medium-width":464,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","medium_large-width":768,"medium_large-height":432,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","large-width":1366,"large-height":768,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","1536x1536-width":1366,"1536x1536-height":768,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","2048x2048-width":1366,"2048x2048-height":768,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst-826x465.png","card_image-width":826,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/businessAnalyst.png","wide_image-width":1366,"wide_image-height":768}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<p><strong>Operations Dashboard: <\/strong>Using operations dashboard containing enriched feature layers, we created easy dynamic access to the status of any structure, the value of the damaged structures, the affected population and much more.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":581702,"id":581702,"title":"Operations dashboard","filename":"opsdashboard.gif","filesize":1179264,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\/opsdashboard","alt":"","author":"8452","description":"","caption":"","name":"opsdashboard","status":"inherit","uploaded_to":581342,"date":"2019-08-07 17:09:20","modified":"2019-08-07 17:09:55","menu_order":0,"mime_type":"image\/gif","type":"image","subtype":"gif","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":600,"height":320,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard-213x200.gif","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","medium-width":464,"medium-height":247,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","medium_large-width":600,"medium_large-height":320,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","large-width":600,"large-height":320,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","1536x1536-width":600,"1536x1536-height":320,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","2048x2048-width":600,"2048x2048-height":320,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","card_image-width":600,"card_image-height":320,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/opsdashboard.gif","wide_image-width":600,"wide_image-height":320}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"<h1><strong>Summary: <\/strong><\/h1>\n<p>Using deep learning, imagery and data enrichment capabilities in the ArcGIS platform, we can quickly distinguish damaged from undamaged buildings, identify the most at-risk populations, and organizations can use this information for rapid response and recovery activities.<\/p>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n<h1><strong>More Resources:<\/strong><\/h1>\n<p><a href=\"https:\/\/pro.arcgis.com\/en\/pro-app\/help\/analysis\/image-analyst\/deep-learning-in-arcgis-pro.htm\">Deep Learning in ArcGIS Pro<\/a><\/p>\n<p><a href=\"https:\/\/doc.arcgis.com\/en\/imagery\/workflows\/resources\/distributed-processing-with-raster-analytics.htm\">Distributed Processing using Raster Analytics<\/a><\/p>\n<p><a href=\"https:\/\/medium.com\/geoai\/building-footprint-extraction-and-damage-classification-8a5458759332\">Details on the model training of the damage assessment\u00a0<\/a><\/p>\n<p><a href=\"https:\/\/www.esri.com\/en-us\/arcgis\/products\/arcgis-notebook-server\">ArcGIS Notebooks<\/a><\/p>\n"}],"authors":[{"ID":8452,"user_firstname":"Vinay","user_lastname":"Viswambharan","nickname":"Vinay Viswambharan","user_nicename":"vinayv","display_name":"Vinay Viswambharan","user_email":"vinayv@esri.com","user_url":"https:\/\/www.esri.com\/arcgis-blog\/author\/vinayv\/","user_registered":"2018-10-04 22:28:54","user_description":"Principal Product manager on the Imagery team at Esri, with a zeal for remote sensing, AI and everything imagery.","user_avatar":"<img data-del=\"avatar\" src='https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/10\/vin4.png' class='avatar pp-user-avatar avatar-96 photo ' height='96' width='96'\/>"},{"ID":6911,"user_firstname":"Rohit","user_lastname":"Singh","nickname":"Rohit Singh","user_nicename":"rsinghesri-com","display_name":"Rohit Singh","user_email":"rsingh@esri.com","user_url":"","user_registered":"2018-03-02 00:19:00","user_description":"Rohit Singh is Director of Esri\u2019s R&amp;D Center in New Delhi, leading the design and development of Geospatial AI capabilities across the ArcGIS platform. He has played a key role in the development of ArcGIS API for Python, ArcGIS Java Engine API, and the Linux enablement of ArcGIS. An alumnus of IIT Kharagpur, Rohit holds an MS in Computer Science with specialization in AI from Georgia Tech.","user_avatar":"<img data-del=\"avatar\" src='https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2025\/08\/RohitSingh_AISummit2025-213x200.jpeg' class='avatar pp-user-avatar avatar-96 photo ' height='96' width='96'\/>"}],"related_articles":"","card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/card5.png","wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/banner1.jpg"},"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v25.9 (Yoast SEO v25.9) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Damage assessment using deep learning in ArcGIS<\/title>\n<meta name=\"description\" content=\"Highlights of how the damage assessment deep learning demo was done by USAA at the UC2019 plenary using the Image Analyst extension in ArcGIS Pro.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Damage assessment using deep learning in ArcGIS\" \/>\n<meta property=\"og:description\" content=\"Highlights of how the damage assessment deep learning demo was done by USAA at the UC2019 plenary using the Image Analyst extension in ArcGIS Pro.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\" \/>\n<meta property=\"og:site_name\" content=\"ArcGIS Blog\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/esrigis\/\" \/>\n<meta property=\"article:modified_time\" content=\"2019-08-11T19:48:59+00:00\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@ESRI\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":[\"Article\",\"BlogPosting\"],\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\"},\"author\":{\"name\":\"Vinay Viswambharan\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/fe6ac8773de48adda0651f5f144d2acf\"},\"headline\":\"Damage assessment using deep learning in ArcGIS\",\"datePublished\":\"2019-08-09T19:10:58+00:00\",\"dateModified\":\"2019-08-11T19:48:59+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\"},\"wordCount\":7,\"publisher\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#organization\"},\"keywords\":[\"ArcGIS Pro\",\"deep learning\",\"image analyst extension\",\"Imagery\",\"machine learning\"],\"articleSection\":[\"Imagery &amp; Remote Sensing\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\",\"name\":\"Damage assessment using deep learning in ArcGIS\",\"isPartOf\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#website\"},\"datePublished\":\"2019-08-09T19:10:58+00:00\",\"dateModified\":\"2019-08-11T19:48:59+00:00\",\"description\":\"Highlights of how the damage assessment deep learning demo was done by USAA at the UC2019 plenary using the Image Analyst extension in ArcGIS Pro.\",\"breadcrumb\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.esri.com\/arcgis-blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Damage assessment using deep learning in ArcGIS\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#website\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/\",\"name\":\"ArcGIS Blog\",\"description\":\"Get insider info from Esri product teams\",\"publisher\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.esri.com\/arcgis-blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#organization\",\"name\":\"Esri\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png\",\"contentUrl\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png\",\"width\":400,\"height\":400,\"caption\":\"Esri\"},\"image\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/esrigis\/\",\"https:\/\/x.com\/ESRI\",\"https:\/\/www.linkedin.com\/company\/5311\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/fe6ac8773de48adda0651f5f144d2acf\",\"name\":\"Vinay Viswambharan\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/10\/vin4.png\",\"contentUrl\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/10\/vin4.png\",\"caption\":\"Vinay Viswambharan\"},\"description\":\"Principal Product manager on the Imagery team at Esri, with a zeal for remote sensing, AI and everything imagery.\",\"sameAs\":[\"https:\/\/www.esri.com\/arcgis-blog\/author\/vinayv\/\",\"https:\/\/www.linkedin.com\/in\/vinayviswambharan\/\"],\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/author\/vinayv\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Damage assessment using deep learning in ArcGIS","description":"Highlights of how the damage assessment deep learning demo was done by USAA at the UC2019 plenary using the Image Analyst extension in ArcGIS Pro.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis","og_locale":"en_US","og_type":"article","og_title":"Damage assessment using deep learning in ArcGIS","og_description":"Highlights of how the damage assessment deep learning demo was done by USAA at the UC2019 plenary using the Image Analyst extension in ArcGIS Pro.","og_url":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis","og_site_name":"ArcGIS Blog","article_publisher":"https:\/\/www.facebook.com\/esrigis\/","article_modified_time":"2019-08-11T19:48:59+00:00","twitter_card":"summary_large_image","twitter_site":"@ESRI","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":["Article","BlogPosting"],"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis#article","isPartOf":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis"},"author":{"name":"Vinay Viswambharan","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/fe6ac8773de48adda0651f5f144d2acf"},"headline":"Damage assessment using deep learning in ArcGIS","datePublished":"2019-08-09T19:10:58+00:00","dateModified":"2019-08-11T19:48:59+00:00","mainEntityOfPage":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis"},"wordCount":7,"publisher":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#organization"},"keywords":["ArcGIS Pro","deep learning","image analyst extension","Imagery","machine learning"],"articleSection":["Imagery &amp; Remote Sensing"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis","url":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis","name":"Damage assessment using deep learning in ArcGIS","isPartOf":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#website"},"datePublished":"2019-08-09T19:10:58+00:00","dateModified":"2019-08-11T19:48:59+00:00","description":"Highlights of how the damage assessment deep learning demo was done by USAA at the UC2019 plenary using the Image Analyst extension in ArcGIS Pro.","breadcrumb":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.esri.com\/arcgis-blog\/"},{"@type":"ListItem","position":2,"name":"Damage assessment using deep learning in ArcGIS"}]},{"@type":"WebSite","@id":"https:\/\/www.esri.com\/arcgis-blog\/#website","url":"https:\/\/www.esri.com\/arcgis-blog\/","name":"ArcGIS Blog","description":"Get insider info from Esri product teams","publisher":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.esri.com\/arcgis-blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.esri.com\/arcgis-blog\/#organization","name":"Esri","url":"https:\/\/www.esri.com\/arcgis-blog\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/","url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png","contentUrl":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png","width":400,"height":400,"caption":"Esri"},"image":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/esrigis\/","https:\/\/x.com\/ESRI","https:\/\/www.linkedin.com\/company\/5311\/"]},{"@type":"Person","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/fe6ac8773de48adda0651f5f144d2acf","name":"Vinay Viswambharan","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/image\/","url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/10\/vin4.png","contentUrl":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/10\/vin4.png","caption":"Vinay Viswambharan"},"description":"Principal Product manager on the Imagery team at Esri, with a zeal for remote sensing, AI and everything imagery.","sameAs":["https:\/\/www.esri.com\/arcgis-blog\/author\/vinayv\/","https:\/\/www.linkedin.com\/in\/vinayviswambharan\/"],"url":"https:\/\/www.esri.com\/arcgis-blog\/author\/vinayv"}]}},"text_date":"August 9, 2019","author_name":"Multiple Authors","author_page":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-pro\/imagery\/damage-assessment-using-deep-learning-in-arcgis","custom_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2019\/08\/banner1.jpg","primary_product":"ArcGIS Pro","tag_data":[{"term_id":42181,"name":"ArcGIS Pro","slug":"arcgis-pro","term_group":0,"term_taxonomy_id":42181,"taxonomy":"post_tag","description":"","parent":0,"count":323,"filter":"raw"},{"term_id":186132,"name":"deep learning","slug":"deep-learning","term_group":0,"term_taxonomy_id":186132,"taxonomy":"post_tag","description":"","parent":0,"count":65,"filter":"raw"},{"term_id":445012,"name":"image analyst extension","slug":"image-analyst-extension","term_group":0,"term_taxonomy_id":445012,"taxonomy":"post_tag","description":"","parent":0,"count":4,"filter":"raw"},{"term_id":115262,"name":"Imagery","slug":"imagery","term_group":0,"term_taxonomy_id":115262,"taxonomy":"post_tag","description":"","parent":0,"count":153,"filter":"raw"},{"term_id":35661,"name":"machine learning","slug":"machine-learning","term_group":0,"term_taxonomy_id":35661,"taxonomy":"post_tag","description":"","parent":0,"count":41,"filter":"raw"}],"category_data":[{"term_id":22931,"name":"Imagery &amp; Remote Sensing","slug":"imagery","term_group":0,"term_taxonomy_id":22931,"taxonomy":"category","description":"","parent":0,"count":768,"filter":"raw"}],"product_data":[{"term_id":36841,"name":"ArcGIS API for Python","slug":"api-python","term_group":0,"term_taxonomy_id":36841,"taxonomy":"product","description":"","parent":36601,"count":151,"filter":"raw"},{"term_id":36561,"name":"ArcGIS Pro","slug":"arcgis-pro","term_group":0,"term_taxonomy_id":36561,"taxonomy":"product","description":"","parent":0,"count":2040,"filter":"raw"}],"primary_product_link":"https:\/\/www.esri.com\/arcgis-blog\/?s=#&products=arcgis-pro","_links":{"self":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/blog\/581342","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/types\/blog"}],"author":[{"embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/users\/8452"}],"replies":[{"embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/comments?post=581342"}],"version-history":[{"count":0,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/blog\/581342\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/media?parent=581342"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/categories?post=581342"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/tags?post=581342"},{"taxonomy":"industry","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/industry?post=581342"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/product?post=581342"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}