Abstract
The development of crop varieties for sustainable food production heavily relies on selecting genotypes with desirable traits such as high yield and tolerance to biotic and abiotic stresses. However, traditional selection procedures in plant breeding are time-consuming, costly, and susceptible to environmental and genetic factors. Leveraging modern digital and information technologies such as artificial intelligence (AI) and big data can support farmers in implementing these techniques, enhancing stakeholder awareness across the agri-food value chain. The AI-CROPBREED project aims to harness the power of AI and image processing to develop sophisticated software capable of accurately estimating early bolting tendency in carrot (Daucus carota L.). By focusing on this specific trait, the project addresses critical challenges within agri-food systems, particularly those related to sustainability, resilience, and economic viability. Through a comprehensive approach, the project seeks to involve stakeholders at every stage of research and development, ensuring that the resulting software meets the diverse needs of the agricultural community. In terms of methodology, the project follows a meticulous approach to enhance carrot breeding practices through the integration of machine learning (ML) and image analysis techniques. Firstly, the project acquires a diverse array of high-quality images of different carrot genotypes. These images undergo rigorous preprocessing aimed at enhancing their quality, reducing noise, and facilitating efficient data organization. Subsequently, ML algorithms are developed, leveraging both image data and bolting date information, potentially incorporating gene expression data of flowering genes for a more comprehensive analysis. Once developed, the algorithm is integrated into an interactive website designed to assist breeders in selecting the most optimal carrots at early plant development stages for further breeding steps. This website serves as a powerful tool for data-driven decision-making and knowledge sharing among stakeholders, effectively breaking down barriers to the adoption of data technologies in agricultural practices.
The project conducts thorough analyses of the benefits and environmental risks linked to big data technologies in the agri-food sector, informing the development of practical advice and innovative solutions to balance their advantages and disadvantages. Ultimately, the AICROPBREED project aims to align stakeholders towards shared objectives, fostering collaboration and driving progress towards a more transparent, sustainable, and resilient food system. Through continuous iteration and refinement of methodologies, it seeks to enhance breeding practices while showcasing the potential of AI-driven innovations to revolutionize agriculture. The AI-CROPBREED project's development of software for predicting early bolting in crops has vast potential applications, impact, and benefits. By enabling the selection of late bolting genotypes resistant to early flowering, it enhances crop resilience and resource efficiency, resulting in reliable harvests and cost savings. Overall, this technology supports sustainable agricultural practices, fostering environmental stewardship, enhancing livelihoods, and improving food security, signifying a substantial advancement in agricultural productivity, resilience, and sustainability through AI-driven innovation.
Project coordinator
Prof. Dr. Meryem Ipek, Bursa Uludag University Agriculture Faculty Horticulture Department, Bursa, Türkiye
Partners
- TURKEY: Prof.Dr. Meryem Ipek, Bursa Uludag University (BUU) Agriculture Faculty Horticulture Department
- TURKEY: Prof.Dr. Mehmet Süleyman Ünlütürk, Yasar University (YU) Software Engineering
- POLAND: Prof. Dr. Dariusz Grzebelus, University of Agriculture in Krakow (UAK) Plant Biology and Biotechnology
- ROMANIA: Dr Violeta Alexandra ION, University of Agronomic Sciences and Veterinary Medicine (USAMV) Research Center for Studies of Food Quality and Agricultural Products
Funding institutions
- BUU and YU: The Scientific and Technological Research Council of Türkiye (TUBİTAK), Turkey
- UAK: The National Centre for Research and Development (NCBR), Poland
- USAMV: Executive Agency for Higher Education, Research, Development and Innovation Funding (UEFISCDI), Romania
Keywords
Artificial intelligence, carrot (Daucus carota L.), crop breeding, bolting, image processing
Project duration
36 months (06.2025-05.2028)
Abstract
The development of crop varieties for sustainable food production heavily relies on selecting genotypes with desirable traits such as high yield and tolerance to biotic and abiotic stresses. However, traditional selection procedures in plant breeding are time-consuming, costly, and susceptible to environmental and genetic factors. Leveraging modern digital and information technologies such as artificial intelligence (AI) and big data can support farmers in implementing these techniques, enhancing stakeholder awareness across the agri-food value chain. The AI-CROPBREED project aims to harness the power of AI and image processing to develop sophisticated software capable of accurately estimating early bolting tendency in carrot (Daucus carota L.). By focusing on this specific trait, the project addresses critical challenges within agri-food systems, particularly those related to sustainability, resilience, and economic viability. Through a comprehensive approach, the project seeks to involve stakeholders at every stage of research and development, ensuring that the resulting software meets the diverse needs of the agricultural community. In terms of methodology, the project follows a meticulous approach to enhance carrot breeding practices through the integration of machine learning (ML) and image analysis techniques. Firstly, the project acquires a diverse array of high-quality images of different carrot genotypes. These images undergo rigorous preprocessing aimed at enhancing their quality, reducing noise, and facilitating efficient data organization. Subsequently, ML algorithms are developed, leveraging both image data and bolting date information, potentially incorporating gene expression data of flowering genes for a more comprehensive analysis. Once developed, the algorithm is integrated into an interactive website designed to assist breeders in selecting the most optimal carrots at early plant development stages for further breeding steps. This website serves as a powerful tool for data-driven decision-making and knowledge sharing among stakeholders, effectively breaking down barriers to the adoption of data technologies in agricultural practices.
The project conducts thorough analyses of the benefits and environmental risks linked to big data technologies in the agri-food sector, informing the development of practical advice and innovative solutions to balance their advantages and disadvantages. Ultimately, the AICROPBREED project aims to align stakeholders towards shared objectives, fostering collaboration and driving progress towards a more transparent, sustainable, and resilient food system. Through continuous iteration and refinement of methodologies, it seeks to enhance breeding practices while showcasing the potential of AI-driven innovations to revolutionize agriculture. The AI-CROPBREED project's development of software for predicting early bolting in crops has vast potential applications, impact, and benefits. By enabling the selection of late bolting genotypes resistant to early flowering, it enhances crop resilience and resource efficiency, resulting in reliable harvests and cost savings. Overall, this technology supports sustainable agricultural practices, fostering environmental stewardship, enhancing livelihoods, and improving food security, signifying a substantial advancement in agricultural productivity, resilience, and sustainability through AI-driven innovation.
Project coordinator
Prof. Dr. Meryem Ipek, Bursa Uludag University Agriculture Faculty Horticulture Department, Bursa, Türkiye
Partners
Funding institutions
Keywords
Artificial intelligence, carrot (Daucus carota L.), crop breeding, bolting, image processing
Project duration
36 months (06.2025-05.2028)