Projects
Automated visual yield estimation in table grape vineyards
The project aimed to introduce and validate the use of machine vision technology in the table grape industry, which had not widely adopted this technology despite its success in other fruit production systems.
The project involved ten table grape producers in north-west Victoria, with selections being based on growing varieties, proximity, and willingness to adopt new technology. The project utilized the GreenView platform by Bitwise Agronomy, which provided imagery and models for shoot, inflorescence, berry, and bunch counting.
The project was structured around several key activities:
The project involved ten table grape producers in north-west Victoria, with selections being based on growing varieties, proximity, and willingness to adopt new technology. The project utilized the GreenView platform by Bitwise Agronomy, which provided imagery and models for shoot, inflorescence, berry, and bunch counting.
The project was structured around several key activities:
- Workshops and Training: Producers were introduced to the technology through workshops and online sessions, which included onboarding documentation and videos. These sessions facilitated networking among producers and provided hands-on experience with the technology.
- Imagery Collection: Imagery was collected at multiple points during the growing season to estimate crop load and assess the efficacy of bunch thinning and trimming practices. However, challenges such as disease pressure and weather conditions affected data collection.
- Validation and Feedback: ATGA staff conducted on-ground validation of the machine vision data and collected feedback from producers. Despite some challenges, the feedback was positive, with expectations of labour savings and improved yield estimates.
The project faced several barriers, including issues with uploading large imagery files, slow internet speeds, and adverse weather conditions. However, the project successfully demonstrated the potential benefits of machine vision technology, such as more accurate yield estimates and better management practices.
The project concluded with plans to extend the use of the technology and address the lessons learned. Bitwise Agronomy agreed to extend the lease of the equipment, allowing producers to continue using the technology for the next season.
The project concluded with plans to extend the use of the technology and address the lessons learned. Bitwise Agronomy agreed to extend the lease of the equipment, allowing producers to continue using the technology for the next season.
