Comparative Evaluation of Convolutional and Transformer-Based Detectors for Automated Weed Detection in Precision Agriculture

TL;DR

This study compares convolutional and transformer-based object detectors for early weed detection, analyzing their accuracy and efficiency trade-offs in realistic agricultural scenarios.

Contribution

It provides a comprehensive evaluation of CNN and transformer models on a weed dataset, highlighting their strengths and limitations for precision agriculture.

Findings

CNN detectors achieve high accuracy with lower computational cost.

Transformer models better capture global context but require more resources.

Results guide model choice based on efficiency and accuracy needs.

Abstract

This paper presents a comparative evaluation of convolutional and transformer-based object detection architectures for early weed detection in realistic scenarios. Representative models from each paradigm are considered, including YOLOv26-nano, a recent variant of the YOLO family, and transformer-based approaches such as RTDETR and RF-DETR. Experiments were conducted on the GROUNDBASED_ WEED dataset, allowing performance to be evaluated in terms of detection accuracy and computational efficiency using metrics such as precision, recall, average precision, and inference speed. The results highlight a clear trade-off between efficiency and contextual modeling: CNN-based detectors achieve high performance at a lower computational cost, while transformer-based approaches offer better global context capture at the expense of higher resource demands. These results provide practical criteria for model selection in precision agriculture applications.