Pixel-Position Gain Filtering for Image Processing Data Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing image processing systems face challenges in managing the large data requirements and computational resources needed for high-accuracy image filtering, particularly due to the exponential increase in data with the number of filters and varying imaging conditions, leading to storage and calculation capacity issues.
Innovation Solution
An image processing apparatus that performs multiple filtering processes with gain adjustments based on target frequency characteristics determined by pixel position, using a filtering process unit that applies filters and gains to acquire processed image data, and a gain specifying unit that determines gains from a gain table associated with pixel positions, reducing the data amount and computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filters are prepared for all combinations of individual imaging conditions, then filtering accuracy is improved, but data amount becomes enormous
Solution Approach 1:
The patent segments the filtering process into multiple stages: first applying a basic filter to all pixels, then selectively applying additional filters only to specific pixel positions (e.g., peripheral regions) where higher accuracy is needed. This segmentation allows the system to achieve high filtering accuracy for critical regions while avoiding the exponential data increase that would result from preparing comprehensive filters for all possible imaging conditions across the entire image.
Solution Approach 2:
The patent implements local quality by applying different filtering strategies to different regions of the image. Central region pixels receive one type of filtering treatment while peripheral pixels receive different treatment. This allows the system to optimize filtering accuracy for specific local regions without requiring enormous data storage for all possible conditions, as each region receives only the filtering necessary for its specific requirements.
2Measurement precision
If filters are prepared for all pixel positions in original image, then filtering accuracy is improved, but processing time increases
Solution Approach 1:
The patent divides the image processing into segments where a first filter is applied to all pixels, and then a second filter is applied only to specific pixel positions that require higher accuracy. This segmentation reduces processing time compared to applying comprehensive filters to all pixels, while still achieving high accuracy for the critical regions that need it most.
Solution Approach 2:
The patent applies partial action by selectively applying additional filtering only to specific pixel positions rather than all pixels. This partial approach achieves sufficient filtering accuracy for the most critical regions (such as peripheral areas with optical aberrations) without incurring the excessive processing time that would result from applying full filtering to every pixel in the image.
3Measurement precision
If point spread function is calculated for all pixel positions, then filtering accuracy is improved, but computation amount increases
Solution Approach 1:
The patent segments the point spread function calculation and filter application process, calculating and applying filters only for specific pixel positions (such as peripheral regions) rather than all pixel positions. This segmentation dramatically reduces the computation amount required while maintaining filtering accuracy for the regions where optical aberrations are most significant.
Solution Approach 2:
The patent implements local quality by focusing computational resources on calculating point spread functions and applying filters only to specific local regions (peripheral pixels) where optical aberrations occur, rather than uniformly processing all pixels. This approach maintains high filtering accuracy for the most problematic regions while minimizing overall computation requirements.
Data Source
AI summary
There are provided an image processing apparatus, an image processing method, a program, and a recording medium capable of compatibly achieving a high-accuracy filtering process and reduction in a necessary storage capacity. An image processing apparatus 35 includes a filtering process unit 41 that performs an image filtering process including a plurality of filtering processes. The filtering process unit 41 applies a filter to processing target data to acquire filter application process data, and applies a gain to the filter application process data to acquire gain application process data, in each filtering process. In each filtering process, the gain applied to the filter application process data is acquired based on a target frequency characteristic of the image filtering process determined according to a pixel position in original image data.