Smart non-contact phenotyping of raspberries and quinces using machine learning methods, hyperspectral and 3D images (AKFen)
Project details
- Status
- Completed
- Period of time
- 01.01.2021. - 31.12.2023.
- Financing
- 299986
- Partners
- 2
- Coordinator
- LZP
- Countries
- Latvia
LZP project No. lzp-2020/1-0353
“Smart Contactless Phenotyping of Raspberry and Quince Using Machine Learning Methods, Hyperspectral and 3D Imaging (AKFEN)”
Project description
The goal of the project is to develop a methodology and tools for non-invasive phenotyping (description and evaluation) of raspberry and Japanese quince yield components based on 3D and hyperspectral imaging and machine learning (ML). To distinguish candidates for cultivars in fruit breeding it is necessary to describe and evaluate the characteristics of several thousand seedlings. This project aims to solve these problems.
On 6th october, researcher Dr.Sc.Comp.Kaspars Sudars and programming engineer Astile Peter attended Microwave Theory and Techniques in Wireless Communications (MTTW'22) and successfully presented the paper on "YOLOv5 Deep Neural Network for Quince and Raspberry Detection on RGB images". In the conference Astile Peter was awarded as the best presenter in the wireless communication section.
