Optical Sensor Model Data Generation for Recognition Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing process for registering model data to optical sensors is labor-intensive and time-consuming, with variations in illumination conditions and camera characteristics leading to inconsistent image generation across different production lines, resulting in uneven recognition accuracy.
Innovation Solution
A method for generating model data by inputting basic model data representing the full-scale shape of an object, performing measurement and imaging processing, and deleting unassociated data to create model data suitable for each optical sensor's conditions, ensuring consistency and accuracy across multiple sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If model data are generated manually for each optical sensor through repeated measurement and correction, then recognition accuracy can be improved, but the time and labor required increases significantly
Solution Approach 1:
The patent applies preliminary action by pre-generating comprehensive model data that includes all possible measurement viewpoints and conditions before actual measurement occurs. This pre-generated model data serves as a template that can be directly matched against actual measurements, eliminating the need for repeated manual measurement and correction cycles while maintaining high recognition accuracy
Solution Approach 2:
The patent uses copying by creating virtual model data representations that can be replicated and applied across multiple optical sensors. Instead of manually measuring and correcting model data for each sensor, the system generates copies of pre-validated model data that adapt to different sensors through automated matching processes, significantly reducing time and labor
2Stability of the object's composition
If the same model data are exported to multiple optical sensors across different production lines, then consistency can be improved, but variations in illumination and camera characteristics cause recognition accuracy to deteriorate
Solution Approach 1:
The patent applies local quality by customizing model data characteristics to match the specific conditions of each optical sensor and production line environment. Instead of using uniform model data everywhere, the system adapts model data properties (such as contrast, illumination characteristics, and camera response) to local conditions while maintaining the core structural consistency of the model, thereby preserving both consistency and accuracy
Solution Approach 2:
The patent uses parameter changes by adjusting model data parameters (such as illumination intensity, contrast levels, and camera calibration values) to compensate for variations in different production line environments. This allows the same base model data to be effectively applied across multiple sensors by dynamically modifying parameters rather than creating entirely separate models
3Loss of information
If model data include comprehensive information from design data, then completeness is improved, but the volume of unassociated information increases, reducing measurement consistency
Solution Approach 1:
The patent applies taking out by extracting only the essential and measurable features from comprehensive design data to create streamlined model data representations. Instead of including all information from design data, the system identifies and extracts only those features that can actually be measured and are relevant for recognition, eliminating redundant information that would reduce measurement consistency
Solution Approach 2:
The patent uses segmentation by dividing model data into distinct functional components: essential measurable features, optional supplementary information, and excluded irrelevant data. This segmentation allows the system to maintain complete and organized model data structures while efficiently managing information volume and ensuring that only appropriate data are used for measurement matching
Data Source
AI summary
To easily generate model data having high recognition accuracy and being consistent with measurement conditions and installation environment of each of optical sensors. Basic model representing a range in which a workpiece can be optically recognized is inputted, and pieces of processing of imaging and measuring the workpiece under the same condition as that in an actual measurement and matching feature data of the workpiece obtained from this measurement with the basic model are executed for a plurality of number of cycles. Then, in the basic model, information is set as unnecessary information where the information cannot be associated with the feature data of the workpiece in all of the pieces of matching processing, or where the number of times or ratio the information cannot be associated is more than a predetermined reference value, or where the information cannot be associated with the feature data in any one of the pieces of executed matching processing. Then, the unnecessary information is deleted from the basic model, and information after each deletion is identified as model data to be registered and is registered to the memory.