ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume V-2-2020
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2020, 475–482, 2020
https://doi.org/10.5194/isprs-annals-V-2-2020-475-2020
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2020, 475–482, 2020
https://doi.org/10.5194/isprs-annals-V-2-2020-475-2020

  03 Aug 2020

03 Aug 2020

NPALOSS: NEIGHBORING PIXEL AFFINITY LOSS FOR SEMANTIC SEGMENTATION IN HIGH-RESOLUTION AERIAL IMAGERY

Y. Feng1,2,3, W. Diao1,2, X. Sun1,2,3, J. Li1,2,3, K. Chen1,2, K. Fu1,2,3, and X. Gao1,2 Y. Feng et al.
  • 1Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, China
  • 2Key Laboratory of Network Information System Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, China
  • 3University of Chinese Academy of Sciences, Beijing, China

Keywords: Deep Learning, Semantic Segmentation, Pixel Weighting Loss, Small-sized Objects, Boundaries, Aerial Imagery

Abstract. The performance of semantic segmentation in high-resolution aerial imagery has been improved rapidly through the introduction of deep fully convolutional neural network (FCN). However, due to the complexity of object shapes and sizes, the labeling accuracy of small-sized objects and object boundaries still need to be improved. In this paper, we propose a neighboring pixel affinity loss (NPALoss) to improve the segmentation performance of these hard pixels. Specifically, we address the issues of how to determine the classifying difficulty of one pixel and how to get the suitable weight margin between well-classified pixels and hard pixels. Firstly, we convert the first problem into a problem that the pixel categories in the neighborhood are the same or different. Based on this idea, we build a neighboring pixel affinity map by counting the pixel-pair relationships for each pixel in the search region. Secondly, we investigate different weight transformation strategies for the affinity map to explore the suitable weight margin and avoid gradient overflow. The logarithm compression strategy is better than the normalization strategy, especially the common logarithm. Finally, combining the affinity map and logarithm compression strategy, we build NPALoss to adaptively assign different weights for each pixel. Comparative experiments are conducted on the ISPRS Vaihingen dataset and several commonly-used state-of-the-art networks. We demonstrate that our proposed approach can achieve promising results.