Depending on the size of the training data gathered, it may be augmented with synthetic plumes in order to avoid overfitting. For detection of temporal activity i.e. tracking of the plumes, the Region Convolutional 3D Network (R-C3D) will be used to extract spatio-temporal features capturing activities, accurately localizing the start and end times of each plume \cite{saenko2017}.

Building Identification

The principles of Photogrammetry will be used to build the 3D environment mapping the buildings on the images to the Lidar data. This step is crucial in identifying the origin of the polluting plume.