Pixel Classification for Noise Reduction and Focus Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in effectively identifying and classifying image pixels, particularly in distinguishing between noise, detail, and edge pixels, which affects image quality and focus determination in images captured by semiconductor imaging devices.
Innovation Solution
A method involving an image analysis process that classifies pixels by determining if they are along edges, in non-edge detail, or in flat areas, using a flow-chart algorithm to calculate threshold and slope values, and employing a 9*Sigma curve to differentiate between noise and image detail, allowing for robust noise reduction and improved image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing is performed to mitigate defects and analyze focus level, then image quality and measurement precision are improved, but device complexity and processing time increase
Solution Approach 1:
The patent segments the image processing task into distinct functional modules: edge detection module, noise analysis module, focus assessment module, and classification module. Each module processes specific aspects of the image independently, improving overall precision while managing complexity through modular architecture. The pixel classification into edge pixels, noise pixels, and detail pixels enables targeted processing strategies.
Solution Approach 2:
The patent performs preliminary edge detection and noise characterization before focus level determination. By pre-identifying edge pixels and analyzing noise properties in advance, the system prepares processing parameters that optimize subsequent focus assessment, improving measurement precision without requiring complex real-time calculations during the actual focus determination.
2Manufacturing precision
If pixels are classified into edge, detail, and noise categories, then image processing accuracy is improved, but computational time and processing complexity increase
Solution Approach 1:
The patent applies different processing criteria and algorithms to different pixel regions based on their characteristics. Edge pixels use edge-specific detection algorithms, detail pixels use local variance analysis, and noise pixels use statistical thresholding. This localized approach improves classification accuracy while reducing overall computational time by avoiding uniform processing of all pixels.
Solution Approach 2:
The patent dynamically adjusts processing parameters based on image characteristics and pixel classification. Threshold values, sigma multiples, and processing intensities are modified according to the specific pixel type and regional properties, enabling accurate classification while optimizing processing speed for each category.
3Reliability
If noise reduction processing is applied to improve image quality, then signal-to-noise ratio increases, but image detail and edge information may be lost
Solution Approach 1:
The patent segments noise reduction processing by pixel type: aggressive denoising for noise pixels, conservative processing for edge pixels, and selective smoothing for detail pixels. This segmentation enables effective noise reduction while preserving critical image information through category-specific processing strategies.
Solution Approach 2:
The patent applies different noise reduction intensities and algorithms to different local regions based on pixel classification. Edge regions receive protection from over-smoothing, detail regions receive targeted processing, and noise regions receive aggressive denoising. This local quality approach maintains signal-to-noise ratio improvement while preserving image detail and edge information.
Data Source
AI summary
A method of classifying pixels in an image is described that includes calculating for each target pixel in the image, a functional value based on a median value of a block of pixels including the target pixel and storing the functional value for each pixel. Pixels in the image are then analyzed to determine if they correspond to edges in the image and if so, are classified as edge pixels. Next the stored functional values are analyzed to define a flat area delimiting function for the image. The stored functional values that do not correspond to edge pixels are then analyzed to define an image detail delimiting function and the non-flat area pixels are classified as being either flat area pixels or detail pixels based on the flat area delimiting function and the detail delimiting function.