Advanced defect detection in blow molded containers using thermal profiling and deep learning

Overview

Our advanced defect detection technology enhances quality control in the blow molding process of thick-walled plastic containers by utilizing real-time thermal profiling combined with deep learning. This innovative solution effectively identifies and differentiates defects such as pearlescence and heat haze, allowing for timely adjustments that improve container quality and minimize defect-related waste. By automating defect detection and classification, it significantly reduces the need for manual inspections, increases production efficiency, and ensures the production of durable, visually appealing containers.

Technical specifications

• Real-time thermal profiling: Captures 360-degree thermal profiles of containers to monitor thermal distribution and detect deviations that cause defects.
• Deep learning integration: Utilizes machine learning algorithms to refine defect detection, accurately identifying thermal variations specific to defects like pearlescence and heat haze.
• Automated feedback system: Provides real-time visual feedback for process adjustments, reducing defect rates and optimizing resource use.
• System adaptation: Configures IR imaging systems for integration into high-speed blow molding lines, ensuring consistent and repeatable defect detection across production batches.

Technology readiness level

This technology is at TRL 6, having been tested and validated in relevant environments. It is ready for implementation on high-speed production lines, demonstrating fast, consistent, and repeatable defect detection.