Super Resolution Mark Recognition Parallel Candidate Search

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

As component mounting densities on circuit boards increase and components miniaturize, reference marks become difficult to accurately recognize due to poor image processing accuracy and longer image processing times caused by multi frame super resolution processing, especially with poor conveyor accuracy and similar patterns in the field of view.

Innovation Solution

An apparatus that performs candidate region search processing in parallel with camera movement to limit the region for multi frame super resolution processing, reducing processing time and improving accuracy by creating a super resolution image only for potential recognition targets within defined candidate regions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multi frame super resolution processing is performed on the entire low-resolution image, then the recognition accuracy of the reference mark is improved, but the image processing time increases and productivity decreases

Engineering Contradiction:
Improvereference mark recognition accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent divides the low-resolution image into multiple candidate regions based on preliminary analysis, and performs super resolution processing only on these candidate regions rather than the entire image. This segmentation approach reduces the processing area and computational load while maintaining recognition accuracy for the reference mark.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing qualities to different regions of the image. Candidate regions that may contain the reference mark receive high-quality super resolution processing, while other regions receive reduced or no processing. This local quality differentiation optimizes the balance between recognition accuracy and processing speed.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If the field of view of the camera is expanded to capture the reference mark, then similar patterns may be captured causing erroneous recognition, but limiting the field of view may miss the reference mark

Engineering Contradiction:
Improvereference mark recognition accuracyVSAvoiderroneous recognition of similar patterns
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent performs preliminary analysis of the low-resolution image to identify candidate regions before performing super resolution processing. This preliminary action allows the system to focus computational resources on regions likely to contain the reference mark while excluding regions containing similar patterns that could cause erroneous recognition.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses feedback from the candidate region selection process to guide the super resolution processing. By analyzing initial image data and using this feedback to define candidate regions, the system iteratively refines the processing focus to accurately locate the reference mark while avoiding similar patterns.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3644702B1Apparatus for performing work on substrate
Publication Date: 2022.07.13 FUJI CORP
  • EP3644702B1 patent drawingFigure 1
  • EP3644702B1 patent drawingFigure 2
  • EP3644702B1 patent drawingFigure 3

AI summary

An apparatus for performing work on a substrate (component mounter) including an image processing device (control device 31) configured to acquire multiple images captured at different positions by performing imaging multiple times after a mark imaging camera (36) has been moved by a moving device (25) such that a reference mark (recognition target) of a circuit board (13) is within a range inside a field of view of the camera, create a super resolution image using multi frame super resolution processing, and recognize the reference mark. The image processing device is further configured to perform candidate region search processing of processing the image captured initially to search for at least one region within the image including a portion that has a possibility of being the reference mark as a candidate region after completion of the initial imaging, parallel with an operation of relatively moving the camera to a subsequent imaging position, perform multi frame super resolution processing on the candidate region searched in the candidate region search processing using the multiple images after completion of final image capturing, create a super resolution image of the candidate region, and recognize the reference mark.