After a large-scale flood in 2018, AI has sped up the process of detecting flooded buildings, allowing emergency personnel to direct their efforts efficiently. A research group from Tohoku University has created a machine learning model that uses news media photos to identify flooded buildings accurately within 24 hours of the disaster.
Normally, the media teams who are often the first on the scene of a disaster to broadcast images to viewers at home, and the research team recognized that this information could be used in AI algorithms. ML and deep learning algorithms are tailored to classify objects through image analysis. For AI and ML to be effective, data is needed to train the model.
From the news broadcast images, AI will specify the areas from the landmark, then SAR (Synthetic Aperture Radar) images can be employed to classify water bodies since microwaves irradiate differently on wet and dry surfaces. A support vector machine (SVM), one of the machine learning techniques, was used to classify buildings surrounded by floodwaters or within non-flooded areas.
The performance of this model estimation resulted in an 80% accuracy. The research group has a plan to explore the applicability of news media databases from past events as training datasets for developing AI Models at present situations to increase the accuracy and speed of classification.