Image Matching Using Two-Stage Edge Pixel Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image measuring apparatuses struggle to accurately determine the positional relation between images when the object's shape varies, as they require a known exact shape of the mark, leading to difficulties in identifying the object's position and posture.
Innovation Solution
A method involving a computer-readable storage medium and an image processing apparatus that obtains and processes first and second image data to specify edge pixels, determine a positional relation in a first degree of accuracy, and then uses usable pixels to achieve a higher second degree of accuracy by excluding inadequate edge pixels, allowing for precise image matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If pattern-matching process using captured image and template image is used to determine positional relation, then the process can be implemented with known mark shape, but the object cannot be identified sufficiently accurately when the object shape varies within individually variability range
Solution Approach 1:
The patent segments the image matching process into two distinct stages: a first degree of accuracy using all edge pixels for initial positioning, and a second degree of accuracy using only selected usable edge pixels for refinement. This segmentation allows the system to first establish a rough match that is adaptable to shape variations, then refine the positional accuracy by selectively using only those edge pixels that correspond between the captured image and template image, thereby resolving the contradiction between adaptability and measurement precision.
Solution Approach 2:
The patent applies local quality by differentiating between usable edge pixels and non-usable edge pixels based on their correspondence between the captured image and template image. Not all edge pixels are treated equally; instead, only those pixels that have valid correspondence relationships are selected for the second degree of accuracy determination. This local differentiation ensures that the positional relation determination is based on reliable, corresponding features while being adaptable to overall shape variations.
2Measurement precision
If all edge pixels are used for determining positional relation, then more data is available for matching, but processing time increases and inaccuracies occur due to inadequate edge pixels
Solution Approach 1:
The patent divides the edge pixel set into two segments: all edge pixels used for the first degree of accuracy determination, and a subset of usable edge pixels used for the second degree of accuracy determination. This segmentation enables efficient processing by first establishing positional relation with comprehensive data, then refining the result with a smaller, more relevant subset of pixels, thereby reducing overall processing time while maintaining high accuracy.
Solution Approach 2:
The patent performs a preliminary determination of positional relation in the first degree of accuracy using all edge pixels before selecting the subset of usable pixels for refinement. This preliminary action establishes a reference framework that guides the subsequent selection of usable pixels, ensuring that the refinement process is efficient and targeted, thus reducing total processing time while achieving high measurement precision.
3Productivity
If pattern-matching process is used with known mark shape, then the process can be implemented efficiently, but it becomes difficult to determine positional relation accurately when object shape varies
Solution Approach 1:
The patent introduces dynamics by making the set of usable edge pixels adaptable rather than fixed. The selection of usable pixels is dynamically determined based on the actual correspondence between the captured image and template image, allowing the system to adjust to shape variations while maintaining processing efficiency. This dynamic adaptation resolves the contradiction between efficiency and precision by enabling the system to flexibly respond to different object shapes.
Solution Approach 2:
The patent changes the parameter of pixel selection criteria based on the matching results. In the first degree of accuracy, all edge pixels are considered. In the second degree of accuracy, the selection criterion changes to include only pixels with valid correspondence relationships. This parameter change allows the system to maintain efficiency through structured processing while achieving high precision by adapting the pixel selection to the actual image content and shape variations.
Data Source
AI summary
A non-transitory computer readable storage medium storing computer readable instructions executable by a computer is provided. The computer readable instructions cause the computer to obtain first image data composing a first image and second image data composing a second image, specify first edge pixels in the first image using the first image and second edge pixels in the second image using the second image data, determine positional relation between the first image and the second image in a first degree of accuracy, based on positional relation between the first edge pixels and the second edge pixels, determine first usable pixels among the first edge pixels and the second usable pixels among the second edge pixels, and using the first usable pixels and the second usable pixels, determine positional relation between the first image and the second image in a second degree of accuracy being higher than the first degree.