About this document

Lightweight Multispectral Weed Segmentation by ياسين حمزوي is a document available to read on EtoBox.

The document presents a lightweight transformer-CNN hybrid model for efficient crop-weed segmentation in precision agriculture, utilizing multispectral imagery (RGB, NIR, RE) to enhance accuracy and robustness. Evaluated on the WeedsGalore dataset, the model achieved a mean Intersection over Union (mIoU) of 78.88%, outperforming traditional RGB-only models by 15.8 percentage points while maintaining a compact architecture of only 8.7 million parameters. The proposed approach aims for real-time deployment on

Author

ياسين حمزوي

Language

EN