Background

Quality inspection is a critical step in industrial food production. Traditional inspection methods rely on manual judgment, which is inefficient, labor-intensive, and prone to missed defects. While 2D machine vision has improved production speed and efficiency, it falls short in measuring biscuit thickness and can miss fine cracks or slight misalignments.

Solution

A 3D camera captures point cloud images of the product in a non-contact manner. The point cloud data is visualized with high-color mapping to generate a 3D model of the product. Height detection and edge detection tools are then used to identify and analyze damage to the biscuits.