CNN Absorbent Article Inspection for In-Line Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current quality control systems for manufacturing disposable absorbent articles are limited by human-generated inspection algorithms, which constrain the speed and scope of in-process inspection operations, making it difficult to conduct sophisticated inspections and predictively maintain manufacturing equipment.
Innovation Solution
The implementation of a method using a convolutional neural network-based inspection algorithm in conjunction with sensors, such as optical or acoustic sensors, to analyze images or sound recordings of absorbent articles and manufacturing machine components, allowing for automatic detection of defects and control actions like rejection or machine adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If human-generated inspection algorithms are used, then the system is easier to implement and understand, but the speed and scope of inspection operations are constrained
Solution Approach 1:
The patent replaces human-generated inspection algorithms with an artificial intelligence-based inspection algorithm that uses machine learning models (such as convolutional neural networks) to automatically analyze sensor data. This substitution enables the system to process inspection data much faster and handle more complex inspection tasks without requiring manual algorithm development, thereby resolving the contradiction between ease of implementation and inspection speed.
2Device complexity
If human-generated inspection algorithms are used, then the system structure is simpler, but the scope of inspection operations is limited
Solution Approach 1:
The patent employs AI algorithms that can dynamically adjust inspection parameters and adapt to different inspection scenarios through machine learning. The system can automatically learn from training data and adjust its inspection criteria, enabling it to handle a broader scope of inspection tasks including defect detection, dimensional measurement, and quality assessment across various absorbent article types without requiring complex manual reconfiguration.
3Device complexity
If traditional sensor technology is used, then the equipment is simpler and cheaper, but sophisticated in-process inspection and predictive maintenance cannot be conducted
Solution Approach 1:
The patent integrates multiple sensor types (optical sensors, acoustic sensors, and other transducers) into a unified inspection system that performs both product inspection and equipment condition monitoring. The AI-based processing unit analyzes data from all sensors to conduct sophisticated inspections of absorbent articles and simultaneously performs predictive maintenance by detecting anomalies in manufacturing equipment, thereby achieving multi-functionality that resolves the contradiction between equipment simplicity and inspection capability.
4Ease of operation
If manual inspection and control actions are used, then the system is easier to control, but productivity and response time are reduced
Solution Approach 1:
The patent implements an automated control system where the AI-based inspection algorithm automatically analyzes sensor data and triggers appropriate control actions without human intervention. The system can automatically reject defective absorbent articles, adjust manufacturing parameters, and alert operators to equipment issues, enabling the production line to maintain high throughput while ensuring quality control. This self-service capability resolves the contradiction between ease of control and manufacturing productivity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables more accurate and efficient inspection of absorbent articles and predictive maintenance of manufacturing equipment, improving product quality and reducing downtime by automating complex inspection tasks and early detection of potential failures.
Implementation Method 1
providing a sensor, preferably selected from the group consisting of an optical sensor, thermal sensor, and combinations thereof
Implementation Method 2
providing a sensor, preferably selected from the group consisting of an optical sensor, thermal sensor, and combinations thereof; providing a reference database, wherein the reference database comprises a plurality of pre-processed images of disposable absorbent articles
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for inspecting disposable absorbent articles, the method comprising: providing a controller comprising an inspection algorithm with a convolutional neural network; providing a sensor, preferably selected from the group consisting of an optical sensor, thermal sensor, and combinations thereof; the sensor being in data communication with the controller; providing a reference database, wherein the reference database comprises a plurality of pre-processed images of disposable absorbent articles and/or components thereof that have been labeled with one or more of: a feature or element of the article and/or component thereof; a non-acceptable defect within a feature or element of the article and/or component thereof; an acceptable defect within a feature or element of the article and/or component thereof; an acceptable faultless feature or element of the article and/or component thereof; and combinations thereof; and the method comprising the steps of: training the convolutional neural network through a plurality of iterations or epochs; optionally validating the training, wherein the training step is repeated until a mean average precision score (MAP) of at least 0.9 is attained according to the formula: MAP=∑q=1QAPqQ where Q is the number of queries; advancing one or more substrates through a converting process along a machine direction to form an array of absorbent articles; creating a plurality of images of said array of absorbent articles with the sensor; transmitting the images from the sensor to the controller; determining at least one of the properties for the labeled component with the inspection algorithm; cutting the substrate into discrete articles; and based on the determination of the least one of the properties for the labeled component, executing a control action, wherein the control action is selected from the group of automatic rejection of one or more articles, automatic machine-setting adjustment, a warning signal for machine maintenance scheduling, and a machine stop command.