Image Processing Apparatus Distortion Correction via Dual Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies fail to accurately generate position data for reference points in photographed images due to lens distortion and noise, leading to incorrect position information.
Innovation Solution
A computer-readable medium stores a program that acquires photographed image data of regularly arranged reference points, analyzes relative positional relations using two methods, determines and excludes inconsistent reference points, and generates position data for the remaining points, ensuring accurate position information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If position information is generated using simple image processing, then processing speed is improved, but position accuracy deteriorates due to lens distortion and noise
Solution Approach 1:
The patent divides the position recognition process into multiple independent methods: a first method using template matching and a second method using feature point detection. Each method processes the image differently, and their results are compared to identify and exclude erroneous position information, thereby improving accuracy without significantly increasing processing time.
Solution Approach 2:
The patent implements a feedback mechanism where the results from the first and second mapping methods are compared against each other. When discrepancies are detected between the two methods, the system identifies the inconsistent reference point as erroneous and excludes it from the final position data, ensuring high accuracy while maintaining efficient processing.
2Measurement precision
If multiple mapping methods are used to verify position information, then position accuracy is improved, but device complexity increases
Solution Approach 1:
The system uses the reference points themselves to verify each other's positions. By comparing the relative positional relationships detected by two different mapping methods, the system automatically identifies and excludes erroneous reference points without requiring external validation, thereby improving accuracy while keeping the processing framework relatively simple.
3Quantity of substance
If all detected reference points are included in position data, then data completeness is improved, but reliability deteriorates due to inclusion of erroneous points from noise
Solution Approach 1:
The patent extracts and excludes erroneous reference points from the final position data by comparing results from two different mapping methods. Reference points that show inconsistent relative positional relationships between the two methods are identified as noise-induced errors and are removed, ensuring that only reliable position information is included in the output.
Data Source
AI summary
A non-transitory computer-readable medium stores a program. The program causes the computer to: acquire photographed image data which is generated by photographing, in an imaging apparatus, a photographed object indicating multiple reference points; acquire pieces of position information which indicate positions of the multiple reference points in the photographed image; map pieces of relative information to the respective reference points by analyzing relative positional relations between the multiple reference points in accordance with a first method; map the pieces of relative information to the respective reference points by analyzing relative positional relations between the multiple reference points in accordance with a second method; and determine a certain reference point as a reference point to be excluded, in a case that a first mapping relation is different from a second mapping relation.