The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B2-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B2-2022, 593–599, 2022
https://doi.org/10.5194/isprs-archives-XLIII-B2-2022-593-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B2-2022, 593–599, 2022
https://doi.org/10.5194/isprs-archives-XLIII-B2-2022-593-2022
 
30 May 2022
30 May 2022

ACTIVE REINFORCEMENT LEARNING FOR THE SEMANTIC SEGMENTATION OF IMAGES CAPTURED BY MOBILE SENSORS

M. Jodeiri Rad and C. Armenakis M. Jodeiri Rad and C. Armenakis
  • Geomatics Engineering, GeoICT Lab, Department of Earth and Space Science, Lassonde School of Engineering, York University, Toronto, Ontario, Canada

Keywords: Semantic Segmentation, Active Learning, Reinforcement Learning, Deep Query Network, Deep Neural Network

Abstract. In recent years, various Convolutional Neural Networks (CNN) have been used to achieve acceptable performance on semantic segmentation tasks. However, these supervised learning methods require an extensive amount of annotated training data to perform well. Additionally, the model would need to be trained on the same kind of dataset to generalize well for other tasks. Further, commonly real world datasets are usually highly imbalanced. This problem leads to poor performance in the detection of underrepresented classes, which could be the most critical for some applications. The annotation task is time-consuming human labour that creates an obstacle to utilizing supervised learning methods on vision tasks. In this work, we experiment with implementing a reinforced active learning method with a weighted performance metric to reduce human labour while achieving competitive results. A deep Q-network (DQN) is used to find the optimal policy, which would be choosing the most informative regions of the image to be labelled from the unlabelled set. Then, the neural network would be trained with newly labelled data, and its performance would be evaluated. A weighted Intersection over Union (IoU) is used to calculate the rewards for the DQN network. By using weighted IoU, we target to bring more attention to underrepresented classes.