Packaging Machine Optical Inspection With Self-Calibrated Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current packaging machines require labor-intensive and time-consuming manual processes to set up tolerance intervals for automatic good or bad product detection, limiting accuracy and increasing false detection rates, which decreases efficiency and increases costs.
Innovation Solution
A packaging machine system that uses an optical sensor and computing unit to determine image feature values and frequency distributions during an initialization phase, allowing for automated classification of products as good or bad based on probability calculations during regular operation, reducing the need for manual setup and improving detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual tolerance interval setting is used for product classification, then detection accuracy can be improved through careful adjustment, but setup time and operational complexity increase significantly
Solution Approach 1:
The system performs self-calibration by automatically determining tolerance intervals through statistical analysis of measured values during operation. The control unit collects measurement data, calculates frequency distributions, and autonomously sets tolerance intervals without requiring manual intervention or test series, enabling the system to configure itself automatically
Solution Approach 2:
The system performs preliminary statistical analysis during an initialization phase before normal operation begins. By collecting and analyzing measurement data in advance, the system pre-determines tolerance intervals so that during regular operation, classification decisions can be made immediately without time-consuming manual setup or trial-and-error adjustments
2Adaptability or versatility
If manual tolerance interval determination is performed on-site, then detection can be adapted to specific installation conditions, but the process becomes personnel-intensive and costly
Solution Approach 1:
The control unit automatically adapts the detection system to specific installation conditions by performing statistical analysis on actual measurement data collected during operation. The system self-configures tolerance intervals based on the unique characteristics of each installation without requiring expert personnel intervention or complex manual adjustment procedures
Solution Approach 2:
The manual mechanical process of tolerance interval setting by personnel is replaced with an automated computational system. The control unit uses statistical algorithms to automatically determine tolerance intervals, substituting human expertise and manual adjustment with automated data analysis and calculation processes
3Ease of operation
If known detection methods with fixed tolerance intervals are used, then the system is simple to operate, but false detection rate increases and efficiency decreases
Solution Approach 1:
The tolerance intervals are made dynamic rather than fixed. The system continuously collects measurement data and automatically adjusts tolerance intervals based on the statistical distribution of actual product variations. This dynamic adaptation allows the system to maintain high detection reliability while remaining easy to operate, as the intervals automatically optimize themselves without requiring complex manual intervention
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A packaging machine (100) comprises an optical sensor (140) and a processing unit (110). The optical sensor (140) is configured to generate image data (410) of the packaging machine's (100's) output products (125, 135). The processing unit (110) is configured to classify an output product (125, 135) of the packaging machine (100) as a good or defective product. To classify the output product (125, 135), a plurality of values for each image feature (420) of one or more predetermined image features is determined during an initialization operation of the packaging machine (125, 135).Determining the plurality of values involves receiving image data (410) of each work product (125, 135) from an initialization set of work products from the optical sensor (140), extracting the predetermined image features (420) for each work product (125, 135) of the initialization set from the received image data (410) of the respective work product (125, 135), and determining a value for each extracted image feature (420) for each work product (125, 135). To classify the work product (125, 135), a corresponding frequency distribution (430) of the respective plurality of values for each predetermined image feature (420) across all work products (125, 135) of the initialization set is determined. Additionally, image data (410) of the work product (125, 135) to be classified is received.Each image feature (420) is extracted from the received image data (410), and a value is determined for each of the extracted image features (420). A probability that the work product to be classified (125, 135) is a good product is determined. The work product to be classified (125, 135) is classified as a good product if the determined probability is greater than or equal to a predetermined threshold.