A Novel Technique for Robust Training of Deep Networks With Multisource Weak Labeled Remote Sensing Data

Deep learning has gained broad interest in remote sensing image scene classification thanks to the effectiveness of deep neural networks in extracting the semantics from complex data. However, deep networks require large amounts of training samples to obtain good generalization capabilities and are sensitive to errors in the training labels. This is a problem in remote sensing since highly reliable labels can be obtained at high costs and in limited amount. However, many sources of less reliable labeled data are available, e.g., obsolete digital maps. In order to train deep networks with larger datasets, we propose both the combination of single or multiple weak sources of labeled data with a small but reliable dataset to generate multisource labeled datasets and a novel training strategy where the reliability of each source is taken into consideration. This is done by exploiting the transition matrices describing the statistics of the errors of each source. The transition matrices are embedded into the labels and used during the training process to weigh each label according to the related source. The proposed method acts as a weighting scheme at gradient level, where each instance contributes with different weights to the optimization of different classes. The effectiveness of the proposed method is validated by experiments on different datasets. The results proved the robustness and capability of leveraging on unreliable source of labels of the proposed method.

Automated summary

Deep learning has attracted attention in remote sensing image scene classification for its ability to extract semantics from complex data, but requires large training samples and is sensitive to errors in training labels. To address this, combining less reliable labeled data sources with a reliable dataset to generate multisource labeled datasets and using a new training strategy that considers the reliability of each source can effectively improve the robustness and capability of leveraging on unreliable sources of labels.