Semi-MCNN: A Semi-supervised Multi-CNN Ensemble Learning Method for Urban Land Cover Classification Using Sub-meter HRRS Images

Runyu Fan, Ruyi Feng, Lizhe Wang, Jining Yan, Xiaohan Zhang
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Sub-meter high-resolution remote sensing (HRRS) image land cover classification could provide significant help for urban monitoring, management, and planning. Deep learning (DL) based models have achieved remarkable performance in many land cover classification tasks through end-to-end supervised learning. However, the excellent performance of DLbased models relies heavily on a large number of well-annotated samples, which is impossible in practical land cover classification scenarios.
more » ... lly, the training set could contain all of the different land cover types. To overcome these problems, in this paper a semi-supervised multiple-CNN ensemble learning method, namely Semi-MCNN, is proposed to solve the land cover classification problem. Considering the lack of labelled samples, a semi-supervised learning strategy was adopted to leverage large amounts of unlabelled data. In the proposed approach, an automatic sample selection method called an ensembled teacher model dataset generation (EMDG) was adopted to select samples and generate a dataset from large amounts of unlabelled data automatically. To tackle the error-propagation problem, an important strategy was adopted to correct the errors by pretraining on the selected unlabelled data and finetuning on the labelled data. Moreover, the semi-supervised idea together with the multi-CNN ensemble framework were integrated into an end-to-end architecture. This could significantly improve the generalization ability of the semi-supervised model, as well as the classification accuracy. Experiments were conducted on Shenzhen's land cover data (ShenzhenLC) and two other public remote sensing datasets. These experiments confirmed the superior performance of the proposed Semi-MCNN compared to the state-of-the-art land cover classification models.
doi:10.1109/jstars.2020.3019410 fatcat:jtzkx4uhojh5jh6nmnrmgim3ce