Image Recognition System with Motion Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image recognition performance is poor when images are captured with a camera that is shaking or when the recognition object is not centered, leading to blurred or defocused images, especially in peripheral areas of the image capturing range.
Innovation Solution
An image processing system that captures images in series, detects recognizable areas, and integrates recognition results based on movement detection and reliability, using a recognition object condition judging portion, movement detecting portion, sub-area detecting portion, recognition portion, and recognition result integration portion to improve accuracy and stability of image recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If image recognition is performed on peripheral areas of the image capturing range, then the recognition coverage is improved, but the recognition accuracy deteriorates due to camera shake and defocus
Solution Approach 1:
The patent divides the image area into multiple regions (center region and peripheral regions) and applies different recognition strategies to each. The center region uses standard recognition, while peripheral regions use motion-compensated recognition based on detected object movement, thereby maintaining accuracy across different areas.
Solution Approach 2:
The patent dynamically adjusts recognition parameters based on the detected movement amount of the recognition object. When movement is detected, the system modifies the recognition process to compensate for the movement, effectively changing the recognition parameters to maintain accuracy despite camera shake or object motion.
2Reliability
If multiple images are captured and integrated to improve recognition accuracy, then the recognition stability is improved, but the processing time increases
Solution Approach 1:
The patent performs integration processing selectively rather than on all captured images. It integrates recognition results primarily for peripheral regions where motion compensation is applied, while using direct recognition for center regions, thereby reducing overall processing time while maintaining stability improvements where needed.
3Measurement precision
If the recognition object is positioned in the center of the image capturing range, then the recognition accuracy is improved, but the adaptability to different object positions deteriorates
Solution Approach 1:
The patent creates a universal recognition system that handles both center and peripheral regions effectively. By detecting object movement and applying motion compensation, the system achieves center-level accuracy for peripheral regions, making the recognition system adaptable to objects positioned anywhere in the image frame while maintaining high accuracy.
Data Source
AI summary
A recognition object detecting portion detects a recognition object existing in each of images which are captured in series at different time points by a camera. Images of the detected recognition object are stored in a memory. A recognition object condition judging portion determines whether image areas of the recognition object are recognizable. A movement amount detecting portion detects a movement amount of the recognition object by using the image areas of the recognition object stored in the memory. A sub-area detecting portion detects a sub-area to be recognized, by using the image areas of the recognition object read out from the memory, in accordance with the movement amount of the recognition object. A recognition portion performs recognition processing on the sub-areas. A recognition result integration portion integrates recognition results of the sub-areas including recognized characters and/or recognized patterns and the corresponding reliabilities, respectively.