Multi-Resolution Noise Filtering Pipeline for Digital Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-pixel digital cameras face significant increases in circuit scale and cost due to the need for large line memories to efficiently filter low-frequency noises that spread across several dozen pixels, making existing multi-resolution noise filtering methods impractical.
Innovation Solution
An image processing equipment and digital camera implementation that uses a pipelined multi-resolution noise filtering approach with minimal line memories, where reduced images are generated and processed in parallel with the original image, allowing for sequential extraction and subtraction of low-frequency noise components without continuous storage of the original-sized image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If vast line memories are prepared to store image data for low-frequency noise filtering, then noise filtering effectiveness is improved, but circuit scale and cost increase significantly
Solution Approach 1:
The patent segments the image processing into multiple stages: original image storage, reduced image generation, noise component extraction from reduced image, up-sampling of noise components, and subtraction from original image. This segmentation allows using small line memories at each stage rather than one large memory, resolving the contradiction between filtering effectiveness and circuit scale.
Solution Approach 2:
The patent implements a nested processing structure where reduced images are generated from the original image, noise components are extracted from the reduced image, and these noise components are then up-sampled and subtracted from the original image. This nested approach allows sequential processing with minimal memory requirements at each level, avoiding the need for vast line memories while maintaining filtering effectiveness.
2Reliability
If vast line memories are prepared to store image data for low-frequency noise filtering, then noise filtering effectiveness is improved, but cost increases
Solution Approach 1:
By segmenting the processing into sequential stages with small line memories at each stage, the total memory capacity required is dramatically reduced compared to storing entire frames, directly lowering manufacturing cost while preserving noise filtering effectiveness through the multi-resolution processing pipeline.
Solution Approach 2:
The patent creates a reduced copy of the original image for noise analysis purposes, processes this copy to extract noise components, and then applies the results back to the original image. This copying approach allows effective noise filtering without requiring large memories to store and process the full-resolution image continuously, reducing cost.
3Productivity
If reduced images are processed in parallel with original image, then processing efficiency is improved, but synchronization complexity increases
Solution Approach 1:
The patent performs preliminary actions by generating the reduced image in parallel with the original image before noise extraction begins. This preliminary generation of the reduced image allows the subsequent noise extraction and up-sampling processes to proceed efficiently with proper synchronization, as the reduced image is already prepared and aligned with the original image processing timeline.
Data Source
AI summary
An image processing equipment generates a reduced image corresponding to an obtained image in parallel with a processing of storing the generated reduced image in a reduced image storing part, and pipelines a processing of extracting a low-frequency noise component of each of pixels included in the obtained image and a processing of sequentially subtracting the low-frequency being extracted from pixel data corresponding to one of the obtained image stored in the image storing part and an adjusted image generated from the obtained, so as to achieve a pipeline processing of a multi-resolution noise filtering with a few line memories.