Image Processing Unit Dynamic Threshold Noise Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods struggle to accurately determine suitability for image synthesis due to the use of fixed thresholds, leading to errors in noise correction and ghosting in images with varying light and dark portions, resulting in suboptimal image quality.
Innovation Solution
An image processing unit that dynamically sets thresholds based on pixel outputs of reference images, using a threshold table correlating mean pixel values to determine suitable image blocks for synthesis, thereby avoiding fixed threshold limitations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a fixed threshold is used to determine image synthesis suitability, then the determination process is simple and fast, but image synthesis accuracy deteriorates in regions with varying light conditions
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed threshold to a variable threshold that adapts to local image characteristics. The threshold is dynamically adjusted based on the mean pixel output values of image blocks, allowing the determination criteria to change according to lighting conditions in different regions of the image.
Solution Approach 2:
The patent changes the parameter of the threshold value based on the mean pixel output of each image block. By calculating the mean pixel output and using it to determine an appropriate threshold, the system adjusts the synthesis determination parameter to match the local lighting conditions, resolving the contradiction between processing speed and accuracy.
2Manufacturing precision
If a large threshold is set to correct noise in light portions, then noise correction accuracy improves, but motion detection capability in dark portions deteriorates
Solution Approach 1:
The patent applies local quality by treating different regions of the image differently based on their lighting conditions. Each image block is evaluated independently, and the threshold is adjusted according to the mean pixel output of that specific block. This allows optimal threshold selection for both light portions (to correct noise) and dark portions (to detect motion), without compromising either region.
3Object-generated harmful factors
If image blocks with large RGB difference values are excluded from synthesis, then ghosting is reduced, but noise correction effectiveness deteriorates
Solution Approach 1:
The patent changes the threshold parameter dynamically based on the mean pixel output of each image block. In dark regions where motion is more likely, a lower threshold allows synthesis to proceed even with smaller RGB differences, preventing ghosting. In light regions, a higher threshold maintains noise correction effectiveness. This dynamic parameter adjustment resolves the contradiction between ghosting reduction and noise correction.
Data Source
AI summary
An image processing unit includes a memory unit in which continuously captured images including a reference image and a comparative image are stored, an image dividing unit to divide the reference image and the comparative image into image blocks of a predetermined size, a mean value calculator unit to calculate a mean value of pixel outputs in each image block of each of the reference and comparative images, a threshold determining unit to determine a threshold according to a mean value of pixel outputs of an image block of the reference image, and a determiner unit to compare the threshold with a difference value of the mean values of the pixel outputs in the image blocks of the reference and comparative images to be synthesized and determine whether the image blocks of the reference and comparative images are suitable for image synthesis based on a result of the comparison.