Area Image Sensor Self-Triggering for Event Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine vision systems in continuous manufacturing processes often require external triggering to capture entire events, leading to inefficiencies and incomplete image capture due to limited frame frequency, especially when dealing with moving objects.
Innovation Solution
An area image sensor captures and transfers a part of the image frame to a processing unit, which analyzes for predefined events; upon detection, the sensor is reconfigured to capture and transfer a larger part of the image frame for further analysis, eliminating the need for external triggering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If external triggering is used to capture entire events, then complete event capture is achieved, but system complexity and response time increase
Solution Approach 1:
The patent extracts the triggering function from an external source and implements it within the image sensor itself through on-chip logic that detects predefined events directly from incoming image data, eliminating the need for external trigger signals while maintaining complete event capture capability
Solution Approach 2:
The image sensor performs self-triggering by incorporating event detection logic that automatically generates trigger signals based on predefined event patterns detected in the image stream, making the system self-sufficient without external control
2Measurement precision
If larger image parts are transferred for analysis, then event detection accuracy improves, but data transmission volume and processing time increase
Solution Approach 1:
The patent segments the image data transmission by transferring only the specific region containing the detected event to the processing unit, rather than transmitting the entire image frame, thus maintaining detection accuracy while minimizing data transmission volume and time
Solution Approach 2:
The system applies different quality levels to different parts of the image by transmitting full-resolution data only for regions containing events of interest, while other regions are either not transmitted or transmitted at lower resolution, optimizing the balance between accuracy and efficiency
Data Source
Figure 1a~1c
Figure 2a~2h
Figure 3~4
AI summary
The invention relates to a method, the method comprises exposing an image frame of an object to be monitored by an area image sensor, transmitting a part of the exposed image frame to an image data processing device, analysing the part and if the part is detected to comprise a predefined event, said image data processing device is arranged to transmit a trigger signal for triggering said area image sensor so that said area image sensor transmits a bigger part of at least one further image frame to said image data processing device for further analysis, and if not the area image sensor continues to expose at least one further image frame and to transmit a part of said at least one exposed further image frame. The invention further relates to a machine vision system for performing the method and a computer program product comprising instructions to perform the method.