Voting Network for Contour Levee Farmland Segmentation and Classification
This work addresses segmentation challenges for agricultural applications, offering incremental improvements in accuracy for farmland analysis.
The paper tackles the problem of segmenting farmlands with contour levees from high-resolution aerial imagery by proposing an end-to-end trainable network that uses a fusion block with voting mechanisms, achieving an average accuracy of 94.34% and improvements of 6.96% and 2.63% in F1 score compared to state-of-the-art methods.
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.