Real-Time Size and Mass Estimation of Slender Axi-Symmetric Fruit/Vegetable Using a Single Top View Image.

Area of Science:

• Agricultural Engineering
• Computer Vision
• Metrology

Background:

• Accurate estimation of physical attributes like mass and volume is crucial for agricultural product quality assessment.
• Traditional stereo-based methods for volume estimation are complex, requiring multiple cameras or manual intervention, hindering practical application in machine vision systems.

Purpose of the Study:

• To develop a simplified and accurate method for estimating the mass and volume of axi-symmetrical agricultural products.
• To overcome the limitations of existing stereo-based approaches by utilizing a single camera setup.

Main Methods:

• A single camera captures a top-view, 2D image of the agricultural product.
• The mass/volume is calculated by decomposing the product into smaller geometric shapes (e.g., chopped pyramids, elliptical cones).
• The method is designed for axi-symmetrical produce, such as carrots and cucumbers.

Main Results:

• The developed method demonstrated good agreement with analytical and experimental results.
• Weight estimation errors were reported as 95% for carrots and 96.7% for cucumbers.
• The technique proved to be accurate, user-friendly, and rotationally invariant.

Conclusions:

• The single-camera approach offers a practical and efficient solution for estimating the mass and volume of axi-symmetrical agricultural products.
• This method enhances the suitability of machine vision systems in agriculture by simplifying data acquisition and processing.
• The findings support the potential for widespread adoption in quality control and automated agricultural processes.