HierbaNetV1: a novel feature extraction framework for deep learning-based weed identification
Justina Michael, Thenmozhi Manivasagam · 2024
Read HierbaNetV1: a novel feature extraction framework for deep learning-based weed identification by Justina Michael, Thenmozhi Manivasagam (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11623067.txt.
Description
Justina Michael and Thenmozhi Manivasagam build a feature extraction frame that names weeds in crop images, plus a field app on top. Weed calls need the right features from images where the span of the weed shifts wildly. Their 2024 frame, HierbaNetV1, pulls features with split-size filters and makes 3,872 feature maps per image, from fine to coarse. Low and high features mix in the learning.
A new sorghum-weed dataset, built for this work, tests the frame next to pre-trained peers. It won at 98.06 percent accuracy. Component cuts show each piece earns its place. Side datasets across crops and weeds back its spread. The app, HierbaApp, puts the call in the hand of a farmer in the field, live.
Catalog record
| CREATOR | Justina Michael, Thenmozhi Manivasagam |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |