Edge Reliability Selection for Position Estimation in Motion Blur
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems fail to accurately estimate the position and orientation of targets in images with uneven image deterioration, such as motion blur, due to constant reduction rates in image data, leading to degraded positional accuracy.
Innovation Solution
An information processing apparatus generates multiple conversion images at different magnifications, detects edges, computes reliability based on luminance gradient values, and selects the most reliable edges for each region, allowing for accurate comparison with a 3D geometric model to compute position and orientation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a constant reduction rate is applied to image data, then processing efficiency is improved, but positional accuracy of edges deteriorates in regions with uneven image deterioration
Solution Approach 1:
The patent applies different reduction rates to different regions of the image based on local image quality characteristics. The image processing unit divides the image into multiple regions and determines appropriate reduction rates for each region, allowing regions with better image quality to use higher reduction rates while regions with poorer quality maintain lower reduction rates to preserve edge accuracy.
Solution Approach 2:
The patent dynamically adjusts the reduction rate based on detected edge characteristics and image quality metrics. Instead of using a fixed constant reduction rate, the system calculates optimal reduction rates adaptively for different regions and edges, making the processing efficiency and accuracy optimization dynamic rather than static.
2Productivity
If image data is reduced to improve processing speed, then productivity is improved, but reliability of edge detection deteriorates
Solution Approach 1:
The patent changes the reduction rate parameter dynamically based on image characteristics. By adjusting this critical parameter according to local image quality, edge sharpness, and detection requirements, the system optimizes the balance between processing speed and detection reliability for each specific region and edge.
Solution Approach 2:
Different reduction rates are applied to different regions based on their specific characteristics. Regions with clear edges and good image quality can tolerate higher reduction rates, while regions with poor quality or critical edges use lower reduction rates to maintain detection reliability.
3Device complexity
If a single reduction rate is used for the entire image, then device complexity is reduced, but manufacturing precision of edge position estimation deteriorates
Solution Approach 1:
The patent segments the image into multiple regions and applies different reduction rates to each segment. This segmentation allows the system to optimize edge position estimation precision for each region while managing complexity through automated region identification and reduction rate assignment algorithms.
Data Source
AI summary
An apparatus includes an input unit configured to input a captured image of a target, a generation unit configured to generate a plurality of conversion images based on the captured image, a detection unit configured to detect edges in the plurality of conversion images, a determination unit configured to determine reliability of the edges detected in corresponding regions in the conversion images and the captured image, and a selection unit configured to select one of the edges for each region in the captured image based on the determined reliability.