Voting Network for Contour Levee Farmland Segmentation and Classification

TL;DR

This paper introduces an end-to-end trainable voting network that improves farmland segmentation and classification from high-resolution aerial imagery by effectively handling small objects and contextual information.

Contribution

The work presents a novel fusion block with voting mechanisms integrated into a segmentation network, enhancing accuracy over state-of-the-art methods.

Findings

Achieved 94.34% average accuracy.

Improved F1 score by 6.96% and 2.63% over existing methods.

Effectively segments and classifies farmland contours and levees.

Abstract

High-resolution aerial imagery allows fine details in the segmentation of farmlands. However, small objects and features introduce distortions to the delineation of object boundaries, and larger contextual views are needed to mitigate class confusion. In this work, we present an end-to-end trainable network for segmenting farmlands with contour levees from high-resolution aerial imagery. A fusion block is devised that includes multiple voting blocks to achieve image segmentation and classification. We integrate the fusion block with a backbone and produce both semantic predictions and segmentation slices. The segmentation slices are used to perform majority voting on the predictions. The network is trained to assign the most likely class label of a segment to its pixels, learning the concept of farmlands rather than analyzing constitutive pixels separately. We evaluate our method using images from the National Agriculture Imagery Program. Our method achieved an average accuracy of 94.34\%. Compared to the state-of-the-art methods, the proposed method obtains an improvement of 6.96% and 2.63% in the F1 score on average.