High-Speed Food Slicer Image Classification for Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current food slicer machines struggle to detect defects in slices, such as contaminants or deformities, before packaging, especially at high cutting speeds, and existing computer vision systems are inadequate for analyzing all slices in a stack efficiently.
Innovation Solution
A slice classification system that captures high-speed digital images of the cutting area, analyzes specific control regions using computer vision algorithms, and compares them to reference patterns to classify slices as acceptable or defective, enabling real-time decision-making on stack quality and rejection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high cutting speeds (900-2000 cuts/minute) are used in industrial slicer machines, then productivity is improved, but it becomes difficult for computer vision equipment to capture and classify images of all slices in real-time
Solution Approach 1:
The system performs preliminary actions by capturing images of slices during the cutting process itself, before the slices are stacked and packaged. The image capture means are positioned to photograph slices as they are being cut, allowing classification to occur before the slices are grouped into stacks, thereby enabling real-time quality control at high cutting speeds
Solution Approach 2:
The patent replaces manual or post-processing quality inspection with an automated computer vision system that uses optical imaging and digital image processing. The system substitutes mechanical inspection methods with electronic image capture and automated classification algorithms, enabling rapid quality assessment that keeps pace with high-speed cutting operations
2Device complexity
If computer vision equipment inspects portions after packaging, then device complexity is reduced, but measurement precision is insufficient because only top and bottom slices are detectable in stacked arrangements
Solution Approach 1:
The system performs quality inspection before the slices are stacked and packaged, capturing images of individual slices as they are being cut. This preliminary inspection ensures that all slices, not just those visible in the final stack, are examined for defects, thereby improving measurement precision without requiring complex post-packaging inspection equipment
Solution Approach 2:
The system creates digital copies (images) of each slice during the cutting process and performs classification on these copies. This allows comprehensive inspection of all slices without physically manipulating or rearranging the actual slices, maintaining device simplicity while achieving complete defect detection coverage
Data Source
Figure 1
Figure 2
Figure 3A~3C
AI summary
The invention relates to a device for classifying slices by processing images captured at high speed during the cutting of food in the shape of bars, such as, for example, sausage, ham or cheese, after which portions are formed which fall or are in stacks and between which a plastic sheet may or may not be inserted. The invention also comprises the method used for classifying these images, with the aim of being able to detect failures in quality in the product before the packaging thereof.