Algorithms based on machine learning for separation of crop and weeds in images
Paris, 20 May- Within CAVIAR, ImViA and MsEyeTech are developing software that can detect and localize crops and weeds for agricultural applications. Their project is based on multi-spectral (MS) imaging that uses visible and near-infrared (NIR) light, and a variety of techniques and methods for MS image analysis. Most important, it is using segmentation – a crucial image processing step that can separate weeds from crops even in complex images. The final goal is to create MS cameras for agricultural applications that can support more efficient and sustainable food production.
Building on these processes, the project aims to develop MS systems that capture images in the visible and NIR ranges, de-mosaic and de-noise them, segment them into crop and weeds, and then provide localisation information for other machines, whether software applications or agricultural robots.
In creating these MS systems, the project is directly addressing the fundamental requirement for detection and monitoring systems in the growing areas of precision agriculture and automated agricultural operations. With MS imaging having proven its ability to extract important information, the software and techniques developed will enable weeds to be detected quickly and accurately – making a significant contribution to the use of new technologies in feeding the world and protecting natural resources.
About the PENTA programme
PENTA is a EUREKA cluster whose purpose is to catalyse research, development and innovation in areas of micro and nanoelectronics enabled systems and applications. Guided by the Electronic Components & Systems (ECS) Strategic Research and Innovation Agenda (SRIA) four technology layers, four cross-sectional technologies and six ECS key application areas, the PENTA programme enables the development of electronic solutions to help drive the digital economy through the formation of collaborative ecosystems along the ECS value chain. This creates the opportunity for rapid competitive exploitation and a strong impact on European societal challenges. PENTA supports SMEs, large corporations, research organisations and universities to work together in project consortia by facilitating access to funding, fostering collaborative work and creating consortia in areas of mutual industrial and National interest.
PENTA is managed by the Industry Association AENEAS
More on PENTA: http://www.penta-eureka.eu
More on AENEAS: https://aeneas-office.org
About the CAVIAR project
CAVIAR is a RD&I project consortium involving 9 partners from 4 countries. The project partners are: Grass Valley Nederland BV (Project leader), 3DHISTECH Ltd, AMS Sensors Belgium, Adimec Advanced Image Systems BV, EVS Broadcast Equipment Brussels (Belgium), EVS Broadcast Equipment SA, MS EYE TECH, TNO and Université de Bourgogne. Belgium, France, the Netherlands and Hungary Public Authorities are funding the project.
More about CAVIAR: https://caviar-project.org/
Read the full press release here.