This is the git repository for the undergradute Machine Learning final project BBM406. Hacettepe University. Computer Engineering Department

Damage assessment after disasters is an important task for governments and organizations, however It takes a lot of time to evaluate the damage especially if the disaster affected a large area. In this project, we propose a solution for this problem by employing image classification methods on damaged structures. We gathered data from the web and categorized them to build the training dataset. We use deep convolutional neural network(CNN), support vector machine(SVM) and bag-of-visual-words(BoVW) models.

https://drive.google.com/file/d/1evJUiFW6iEQUo_efy8Gfitw75DHwDlfw/view?usp=sharing

https://drive.google.com/file/d/13s3yG8LmumqZlxKNaWQNIKDNPfNE7b8f/view?usp=sharing

https://drive.google.com/file/d/18ZpsVVO3_ORDphOhNusrYzQBbPfHxLdb/view?usp=sharing

par1: https://medium.com/bbm406f19/week-1-damage-assessment-system-for-imagery-data-dasm-3764a3b24871

part2: https://medium.com/bbm406f19/week-2-dasm-dataset-building-ed88890a915b

part3: https://medium.com/bbm406f19/week-3-dasm-dataset-building-and-model-selection-87e3e29931a

part4: https://medium.com/bbm406f19/week-4-dasm-bovw-implementation-35f223ae9192

part5: https://medium.com/bbm406f19/week-5-dasm-enlarging-the-dataset-and-convnet-classification-afada08faf0a

part6: https://medium.com/bbm406f19/week-6-dasm-new-experiments-and-the-final-dataset-53250a6804ea

Since the repo is using SIFT and SURF descriptors from openCV which are now not supported, specific opencv, opencv-contrib and python version should be satisfied.

. Python==3.6.0

. opencv-python==3.4.2.16

. opencv-contrib-python==3.4.2.16

malek-baba, ezgigungor, fnisahopcu