Image Processing Apparatus Position Alignment Using Contrast Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In image processing, regions with blown out highlights (high luminance and low contrast) can lead to erroneous position displacement detection when combining multiple images, causing artifacts and reducing alignment precision.
Innovation Solution
An image processing apparatus and method that divide images into regions, calculate motion vectors, detect specific high-luminance and low-contrast regions, and exclude their motion vectors to improve position alignment accuracy by reducing artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pixel values greater than a prescribed threshold are used to detect blown out highlights, then high-luminance regions can be identified, but regions with low contrast (such as clouds) cannot be distinguished from valid image content, leading to erroneous exclusion of motion vectors
Solution Approach 1:
The patent changes from using a single threshold parameter to using two parameters simultaneously: luminance threshold and contrast threshold. By calculating contrast as (maximum pixel value - minimum pixel value) in each block and comparing it against a contrast threshold, the system can distinguish between high-luminance regions with low contrast (blown out highlights) and high-luminance regions with normal contrast (valid content like clouds). This dual-parameter approach resolves the contradiction by improving detection precision while maintaining reliability of motion vector detection.
2Manufacturing precision
If all high-luminance regions are excluded from motion vector detection, then blown out highlights are handled correctly, but valid regions with high luminance (such as clouds) are incorrectly excluded, causing position alignment errors
Solution Approach 1:
The patent applies local quality by making the exclusion criterion specific to each region rather than applying a global exclusion rule. Each block of pixels is evaluated individually based on its local contrast characteristics. Regions with high luminance and low contrast (blown out highlights) are excluded, while regions with high luminance and normal contrast (clouds) are retained. This localized evaluation approach improves position alignment precision without unnecessarily increasing overall system complexity.
Solution Approach 2:
The system introduces contrast as an additional parameter to the region detection process. By calculating the difference between maximum and minimum pixel values within each block and comparing it against a contrast threshold, the system creates a more nuanced detection mechanism. This parameter change allows the system to differentiate between problematic high-luminance regions and valid high-luminance regions, improving position alignment precision while keeping the detection process computationally manageable.
3Productivity
If motion vectors from high-luminance regions are used for position alignment, then more data is available for alignment, but erroneous motion vectors from blown out highlights contaminate the alignment process
Solution Approach 1:
The patent extracts and excludes motion vectors from specific problematic regions before performing position alignment. By identifying blocks with high luminance and low contrast, the system extracts only the reliable motion vectors from regions that do not meet the exclusion criteria. This selective extraction ensures that erroneous motion vectors from blown out highlights do not contaminate the alignment process, maintaining measurement precision while still utilizing motion vectors from valid regions for efficient image combination.
Solution Approach 2:
The patent introduces contrast calculation as an intermediary step between luminance detection and motion vector selection. This intermediary mechanism acts as a filter that determines whether high-luminance regions should have their motion vectors excluded. By using contrast as an intermediary criterion, the system can selectively include or exclude motion vectors based on regional characteristics, ensuring both efficiency and precision in the image combination process.
Data Source
AI summary
An image processing apparatus includes one or more processors including hardware. The one or more processors are configured to: divide, regarding a plurality of images acquired by capturing images of an imaging subject over time, the respective images into a plurality of regions; calculate, for the respective divided regions, motion vectors by detecting movements of the imaging subject; detect luminances and contrasts of the respective regions and detect specific regions where the detected luminances and contrasts satisfy determination conditions, the determination conditions being that the luminances are equal to or greater than a first threshold and that the contrasts are equal to or less than a second threshold; select the motion vectors to be used by excluding the motion vectors of the detected specific regions from the calculated motion vectors; and generate a combined image by performing position alignment of the plurality of images by using the selected motion vectors.