A Parameter-efficient Convolutional Approach for Weed Detection in Multispectral Aerial Imagery

TL;DR

FCBNet is evaluated on the WeedBananaCOD and WeedMap datasets under both RGB and multispectral modalities, showing that FCBNet outperforms models such as U-Net, DeepLabV3+, SK-U-Net, SegFormer, and WeedSense in terms of mIoU, while also achieving superior computational efficiency.

Abstract

We introduce FCBNet, an efficient model designed for weed segmentation. The architecture is based on a fully frozen ConvNeXt backbone, the proposed Feature Correction Block (FCB), which leverages efficient convolutions for feature refinement, and a lightweight decoder. FCBNet is evaluated on the WeedBananaCOD and WeedMap datasets under both RGB and multispectral modalities, showing that FCBNet outperforms models such as U-Net, DeepLabV3+, SK-U-Net, SegFormer, and WeedSense in terms of mIoU, exceeding 85%, while also achieving superior computational efficiency, requiring only 0.06 to 0.2 hours for training. Furthermore, the frozen backbone strategy reduces the number of trainable parameters by more than 90%, significantly lowering memory requirements.

Figures (6)

• Figure 1: Overview of FCBNet proposed in this work.
• Figure 2: Comparison between ResNet, Swin Transformer, and ConvNeXt blocks.
• Figure 3: ConvNeXt-base architecture. Adapted from: ramos2026multiencoder
• Figure 4: Structure of the Feature Correction Block.
• Figure 5: Smoothing block and head used in FCBNet decoder.