Archive

Downloads, videos and webinars from PhenomUK.net

PhenomUK was launched in February 2019, as a UKRI Technology Touching Life (TTL) Network.

The network brought together the UK phenomics community, funded 11 pilot projects developing and examining phenotyping tools and technologies and ran annual meetings, workshops and webinars on a variety of topics. The TTL project ended on March 31st 2023, as PhenomUK took on the work of the UK Plant and Crop Phenotyping Infrastructure scoping activity. This page provides access to the various materials produced by the initial PhenomUK network and previously available via www.phenomuk.net.

Archive

PhenomUK Community Software Data Requirements report

PDF UKRI Scoping Project
Report mapping the community requirements on Software and Data Authors: Sotirios A Tsaftaris (s.tsaftaris@ed.ac.uk) Valerio Giuffrida (valerio.giuffrida@nottingham.ac.uk) This report aims to provide recommendations of the software provision required to run and support the digital research infrastructure for the PhenomUK community. To have a better idea of the needs of the UK plant research community, we conducted several interviews and collected information on three major aspects: (i) data collection; (ii) software tools; (iii) FAIR data. After several months of gathering insights from researchers via interviews, we found the following points as the current pressing issues the community faces: (i) Manual software pipelines cause bottlenecks in data analysis; (ii) Bespoke code gets forgotten, becoming legacy in a short timeframe; (iii) Data storage is always insufficient as needs grow; (iv) Data curation and sharing is still an issue for many. To address these community needs, we propose a set of recommendations to inform future directions of the PhenomUK Scoping Project.

Adapting Vision Foundation Models for Plant Phenotyping

UKRI Scoping Project
Foundation models are large models pre-trained on tremendous amount of data. They can be typically adapted to diverse downstream tasks with minimal effort. However, as foundation models are usually pre-trained on images or texts sourced from the Internet, their performance in specialized domains, such as plant phenotyping, comes into question. In addition, fully fine-tuning foundation models is time-consuming and requires high computational power. This paper investigates the efficient adaptation of foundation models for plant phenotyping settings and tasks.

EMPHASIS Structure and Proposed Services

PDF UKRI Scoping Project