CVLGIVAug 24, 2020

LULC Segmentation of RGB Satellite Image Using FCN-8

arXiv:2008.10736v19 citations
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of automated land classification from satellite imagery for remote sensing applications, but it is incremental as it adapts an existing FCN method to a specific dataset and task.

The paper tackles land use land cover segmentation of RGB satellite images by applying an FCN-8 model with a non-overlapping grid-based approach, achieving an average accuracy of 91.0% and IoU of 0.84, which significantly outperforms eCognition software at 74.0% accuracy and 0.60 IoU.

This work presents use of Fully Convolutional Network (FCN-8) for semantic segmentation of high-resolution RGB earth surface satel-lite images into land use land cover (LULC) categories. Specically, we propose a non-overlapping grid-based approach to train a Fully Convo-lutional Network (FCN-8) with vgg-16 weights to segment satellite im-ages into four (forest, built-up, farmland and water) classes. The FCN-8 semantically projects the discriminating features in lower resolution learned by the encoder onto the pixel space in higher resolution to get a dense classi cation. We experimented the proposed system with Gaofen-2 image dataset, that contains 150 images of over 60 di erent cities in china. For comparison, we used available ground-truth along with images segmented using a widely used commeriial GIS software called eCogni-tion. With the proposed non-overlapping grid-based approach, FCN-8 obtains signi cantly improved performance, than the eCognition soft-ware. Our model achieves average accuracy of 91.0% and average Inter-section over Union (IoU) of 0.84. In contrast, eCognitions average accu-racy is 74.0% and IoU is 0.60. This paper also reports a detail analysis of errors occurred at the LULC boundary.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes