In Sassenheim, The Netherlands, Condi Food has recently unveiled an advanced inspection system designed for the food industry, utilizing hyperspectral imaging technology to evaluate cherry tomatoes based on their taste. This system employs a vision-based approach, which might seem unconventional at first for assessing a subjective attribute like taste. However, hyperspectral imaging captures detailed spectral information of light, extending beyond visible RGB, to analyze the physical and chemical properties of objects, including their taste potential.
The conventional method for sorting produce such as tomatoes primarily focuses on color and shape to determine ripeness. Condi Food's client, already equipped with an automated sorting system using RGB cameras for this purpose, faced limitations in identifying the most flavorful tomatoes solely based on these criteria. As Jacques van Munster van Heuven, Managing Director of Condi Food, explains, color alone proved insufficient for selecting the desired quality, leading to the development of a system that could also assess taste.
The hyperspectral imaging system introduced by Condi Food integrates with existing sorting mechanisms, incorporating a Specim FX17 near-infrared camera and custom-built halogen lighting to effectively scan and analyze tomatoes. This setup, capable of processing 60 tomatoes per second, provides comprehensive spectral data, which, when processed through Condi's machine vision software, facilitates the sorting of tomatoes based on size, color, and now, taste.
Tomatoes are assessed by the system through spectral imaging, capturing data on chemical composition and other internal characteristics. This information, compared against standards set by a taste panel, is used to create inspection models that assign quality levels to each tomato. The outcomes of these models enable growers to sort their produce more accurately according to taste criteria.
Developing these models presented challenges due to the complex nature of tomatoes and the dynamic changes they undergo post-harvest. Nonetheless, the system's flexibility and the need for customization for different produce types underscore its innovative approach to quality assessment in the food industry.
Source: Vision-Systems