TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Active Learning Strategies Compared for YOLOv8 on Lincolnbeet

3 pointsby MattyMattabout 2 years ago

1 comment

MattyMattabout 2 years ago
TLDR:<p>Machine learning applications in agriculture have an incredible impact on water, herbicides, pesticides, and fertilizer usage! In many cases, a reduction of up to 50% is possible by using ML for precision spraying.<p>Excited to share the impressive results on active learning in agriculture that Igor Susmelj just published!<p>Leveraged active learning to supercharge our YOLOv8 model for lincolnbeet (sugarbeet and weed detection). Boosted mean Average Precision (mAP) by up to 14.6x compared to random image labeling. Active learning reduced annotation costs by up to 77% - perfect for optimizing your annotation budget!<p>It&#x27;s incredible how data selection by using active learning can make such a huge impact!