MACHINE-BASED THREE-DIMENSIONAL (3D) OBJECT DEFECT DETECTION
Patent Information
- Application Number
- DE602019063737
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2018-11-16
- Filing Date
- 2019-11-15
- Publication Date
- 2024-12-18
- Estimated Expiration
- 2039-11-15
AI Technical Summary
Current manual inspection methods for detecting defects in 3D printed objects are highly subjective, leading to high rates of false positives and false negatives, resulting in either rejection of defect-free objects or overlooking defects in flawed ones, which can cause malfunctions or failures during use.
An automated imaging defect detection system using machine learning models to analyze images of 3D printed objects, employing a multi-axis platform, various light sources, and imaging devices to identify manufacturing defects such as layering defects, gross defects, and interface defects without user input.
The system significantly reduces false positive and false negative determinations, ensuring higher accuracy and consistency in defect detection, thereby preventing defective 3D printed objects from being used or shipped, and allowing for timely remanufacturing or repair.