Low-Light Image Noise Removal Using Statistical Filters
Solution Overview
Problem
Conventional image processing methods fail to effectively remove Poisson and photon counting noise and false color noise in low light environments, leading to image quality deterioration and inefficient compression, while also smearing edges and details.
Innovation Solution
An image processing method and system that segment images into unit areas, using statistical information to differentiate between low and high light regions, and employing a spatial hybrid filter to remove Poisson and photon counting noise, and a statistical domain temporal filter to remove false color noise, while maintaining edge and detail information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional spatio-temporal filtering method is used to remove noise, then noise removal is achieved, but image quality deteriorates due to smearing in areas without noise
Solution Approach 1:
The patent applies different filtering strategies to different regions of the image based on local characteristics. It identifies and processes Poisson noise, photon counting noise, and false color noise separately using region-specific methods, thereby removing noise while preserving edge and detail information in areas without noise, thus avoiding smearing artifacts.
Solution Approach 2:
The patent segments the image processing task into distinct components: separating Poisson noise removal, photon counting noise removal, and false color noise removal. Each type of noise is detected and processed independently using appropriate statistical methods, allowing precise noise removal without affecting other image regions.
2Object-affected harmful factors
If conventional filtering method is used to remove Poisson and photon counting noise, then noise removal is attempted, but detection accuracy is insufficient leading to ineffective filtering
Solution Approach 1:
The patent replaces conventional mechanical filtering approaches with statistical domain processing. It uses statistical information (mean and variance calculations) to detect and remove Poisson and photon counting noise, achieving more accurate noise detection and removal compared to traditional spatial or temporal filters.
Solution Approach 2:
The patent changes the processing domain from spatial or temporal filtering to statistical domain processing. By calculating statistical parameters (mean and variance) of image data and using these to detect noise characteristics, the method achieves superior noise removal accuracy for Poisson and photon counting noise.
3Object-affected harmful factors
If conventional filtering method is used, then single type of noise is removed, but multiple types of noise (Poisson, photon counting, and false color) remain
Solution Approach 1:
The patent creates a universal noise removal system that handles multiple types of noise (Poisson noise, photon counting noise, and false color noise) within a single processing framework. The system adaptively identifies and applies appropriate removal methods for each noise type, making it versatile for low light environment image processing.
Solution Approach 2:
The patent employs dynamic noise removal by detecting motion pixels and adapting the filtering process accordingly. For false color noise, it uses temporal filtering on stationary pixels while preserving motion information, making the noise removal process adaptive to different image regions and noise types.
4Productivity
If image is compressed from various formats or transmitted in low light environment, then image transmission and storage are achieved, but noise deteriorates image quality and compression efficiency
Solution Approach 1:
The patent applies noise removal as a preliminary processing step before image compression. By removing Poisson noise, photon counting noise, and false color noise from the original image data, it improves the quality of input data for compression, thereby enhancing overall compression efficiency and reducing the bitrate required for acceptable image quality.
Data Source
AI summary
Disclosed are an image processing method capable of effectively removing noise contained in an image photographed in a low light environment by using statistical information, and a system thereof. There is provided an image processing method for an image consecutively inputted per frame unit, the method comprising: a step of segmenting the image into unit areas comprising a predetermined number of pixels; a first step of dividing each unit area into a low light region or a high light region by using the brightness of image data contained in said each unit area; a second step of outputting statistical information from image data contained in at least one unit area divided into the low light region, and detecting and removing Poisson and photon counting noise on the basis of the statistical information, the statistical information being the mean of the unit area; a third step of detecting a motion pixel from the image data; and a fourth step of detecting and removing false color noise from the image data on the basis of the detected motion pixel and the statistical information.