AI-based computer vision is a field of artificial intelligence that enables machines to see, interpret and make decisions based on visual data—just like humans, but with greater speed, precision and consistency.
In the agri-food sector, speed and accuracy are critical to maintaining competitiveness. However, many processes are still carried out manually, leading to inefficiencies, errors and high operating costs. Two clear examples are pepper counting and supervision in production lines for canned goods.
Case 1: Manual pepper counting
Traditionally, the counting and classification of peppers is performed manually—a slow, costly method prone to human error. This results in inconsistent quality, limits the ability to process large volumes and increases the workload. With Tupl Vision, this process is automated through AI-powered computer vision, boosting accuracy from 75% to 90% and multiplying by five the number of samples analysed. The outcome: a drastic reduction in processing times, lower operational burden and more consistent, reliable sorting.
Case 2: Lack of visibility on the production line
In the manufacturing of canned goods packed in glass jars with aluminium lids, the lack of control led to errors such as incorrect quantities, excess liquid and faulty lids—issues that compromised perceived quality and led to returns. Tupl Vision detects and corrects these errors in real time, ensuring every jar meets quality standards. This minimises returns, guarantees full traceability and optimises efficiency, strengthening consumer trust and reducing reprocessing costs.
Key benefits of the Tupl Vision system
The main improvements achieved with the implementation of this technology include increased precision and consistency in classification, reduced processing times and operational costs, full production control and traceability, and an overall improvement in perceived quality and customer satisfaction.
