Please use this identifier to cite or link to this item:
|Title:||Deep learning for aerial image analysis||Authors:||Lim, Benjamin Hong Siong||Keywords:||DRNTU::Engineering::Computer science and engineering||Issue Date:||2019||Abstract:||This project explores the topic of deep learning and how to implement it onto image analysis, namely image classification and image detection. The topic of focus would be the classification and detection of wildfires as images could be captured from drones to fulfill the aerial image segment of the project. As climate change is a growing issue today, we bring our attention to one of the causes, global warming due to deforestation by fire. This can be caused both naturally due to high temperatures or by men using irresponsible slash and burn methods for farming. As current ways of detecting wildfires is still lacking, the drive of the project would be to explore wildfire detection through image recognition. The first half of the project explores object classification and how to classify an image as one that contains a wildfire and one that does not. In the second half, upon classifying the image as a wildfire, an image detection algorithm will be ran to “locate” the fire. This project also experiments on different models to see which ones produces the best results for both techniques.||URI:||http://hdl.handle.net/10356/76920||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||SCSE Student Reports (FYP/IA/PA/PI)|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.