Event-Based Surface Inspection for Moving Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image sensors struggle to detect defects on a homogeneous surface of a moving object due to motion blur, as they require fast frame rates that preclude sufficient exposure times, and event-based cameras alone are insufficient for defect detection without motion compensation.
Innovation Solution
An event-based camera system captures events and compensates for object movement by calculating each event back to a reference point, generating a stationary image from corrected events, and uses machine learning for defect detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional CMOS image sensor is used with sufficient exposure time to capture defects, then image quality is improved, but motion blur occurs due to object movement during exposure
Solution Approach 1:
The patent transitions from static frame-based capture to dynamic event-based capture, where each pixel independently detects intensity changes and generates events asynchronously. This dynamic approach allows the system to adapt to object movement by capturing only relevant changes, eliminating motion blur while maintaining defect detection accuracy.
Solution Approach 2:
The patent replaces the mechanical shutter and fixed exposure time mechanism of conventional cameras with an event-driven computational approach. Instead of using a physical shutter to control exposure, the system uses algorithmic processing of asynchronous events to reconstruct images, substituting mechanical timing control with computational methods that can handle moving objects.
2Speed
If a line scan camera is used to achieve high frame rates and reduce motion blur, then exposure time is reduced, but the field of view becomes very limited
Solution Approach 1:
The patent segments the image capture process into independent pixel-level events rather than capturing the entire field of view simultaneously. Each pixel operates independently, detecting intensity changes and generating events asynchronously. This segmentation allows the system to maintain a wide field of view while achieving high temporal resolution, as each pixel effectively captures at its own optimal rate based on local scene dynamics.
3Measurement precision
If events are accumulated over an extended period to improve defect detection, then sensitivity is improved, but motion blur occurs as events scatter across the field of view
Solution Approach 1:
The patent applies preliminary motion compensation by calculating the object's movement trajectory and using this information to correct event positions before accumulation. By predicting and compensating for motion in advance, the system maintains spatial accuracy of events while allowing them to be accumulated over extended periods, thereby improving defect detection sensitivity without scattering events across the field of view.
Solution Approach 2:
The patent implements feedback through continuous tracking of object movement and real-time adjustment of event positions. The system monitors the relationship between object position and event generation, using this feedback to dynamically correct spatial coordinates of events during accumulation, ensuring that events remain accurately attributed to their source locations even over extended time periods.
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
Enables accurate detection of even small defects on moving objects with continuous data acquisition and high precision, preventing defects in battery production by inspecting films for particles as small as half the film thickness.
Implementation Method 1
Each pixel element independently checks whether it detects a change in intensity. Only in this case is image information generated and output, or read out, and only by that specific pixel.
Implementation Method 2
The events are corrected according to the time elapsed since a reference point and the movement that occurred during that time; in other words, the motion is reversed based on the elapsed time in order to compensate for motion blur.
Data Source
Figure 1~2
Figure 3~4
AI summary
A method is specified for detecting defects (17) on a homogeneous surface (16) of an object (14) during movement of the object (14), wherein an image of the homogeneous surface (16) is recorded and the image is evaluated to determine whether it contains defects (17). The image is recorded using an event-based image sensor (20) which uses a plurality of pixel elements to detect events of varying intensity. The recording takes place over a time interval within which the homogeneous surface (16) continues to move over at least some pixel elements of the image sensor (20). The events are corrected according to a time elapsed since a reference time and the movement occurring during the elapsed time. The corrected events are collected in an image which is then evaluated with regard to the defects (17).