Using Deep-Learning for Weed Identification from UAV-Imagery

Dharmendra Saraswat, associate professor of agricultural and biological engineering, and Varun Aggarwal, graduate student in agricultural and biological engineering at Purdue University recently presented a webinar titled How Good is Deep-Learning for Weed Identification Using UAV-Imagery.
Deep-learning methods are showing a great promise for identifying objects of interest based on benchmark test datasets. This presentation discusses results from a study assessing suitability of a few deep-learning models for identifying weeds based on real-field imagery data acquired using unmanned aerial vehicle (UAV).
This video is part of a weekly webinar series featuring experts in data science and digital agriculture at Purdue University that kicked off on Feb. 4 at 12:30 p.m. ET. Each Thursday through May 13, the Data Driven Agriculture webinars will explore ways digital agriculture and data science can impact agriculture today and into the future. For more information, visit ag.purdue.edu/digital-ag-reso...

[@sophiez7952]

you labeled all the ground truth by yourself by hand or by software?