Endoscope Image Processor Nonlinear Gradation Noise Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In electronic endoscope devices, contrast of images decreases in areas with gentle tone curve changes, leading to image collapse in dark portions, and brightness conversion often emphasizes noise in regions with rapid brightness changes, causing issues like scattered black dots in bright areas.
Innovation Solution
A processor for endoscopes with an image processing unit that performs nonlinear gradation conversion, using a preprocessing unit to set reference upper and lower limit characteristic lines and calculate adjustment values based on pixel variation, outputting emphasis-processed pixel values to suppress noise in flat regions without collapsing dark portions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If nonlinear gradation conversion is performed to emphasize contrast in captured images, then visibility of living tissue structures is improved, but noise is emphasized in flat portions causing scattered black dots
Solution Approach 1:
The patent applies different processing characteristics to different regions of the image based on local pixel value variations. Flat portions (with low variation) receive processing that suppresses noise emphasis, while non-flat portions (with high variation) receive processing that maintains contrast enhancement. This is achieved by calculating the standard deviation of pixel values in local regions and adjusting the gradation conversion accordingly.
Solution Approach 2:
The patent performs preliminary calculation of pixel value variations in local regions before applying the gradation conversion. By pre-identifying flat portions through variance calculation, the system can prepare appropriate processing parameters to prevent noise emphasis before the actual contrast enhancement is applied.
2Illumination intensity
If tone curve adjustment is performed to brighten dark portions, then visibility of dark areas is improved, but image collapse occurs in portions with gentle tone curve changes
Solution Approach 1:
The patent adjusts the tone curve application based on local image characteristics. In regions with gentle tone curve changes (identified through pixel variation analysis), the processing is modified to prevent collapse while still improving visibility. This allows different parts of the image to have optimized processing parameters suited to their local characteristics.
Solution Approach 2:
The patent dynamically adjusts processing parameters based on local pixel value distributions. Rather than applying a fixed tone curve transformation, the system adapts the conversion characteristics according to the local variance and brightness distribution, enabling flexible control over brightness enhancement while maintaining image integrity.
3Illumination intensity
If global brightness conversion is applied to the entire image, then overall image brightness is improved, but noise is emphasized in specific regions causing scattered black dots
Solution Approach 1:
The patent divides the image into local regions and processes each region independently based on its characteristics. By segmenting the image processing into local operations rather than a single global transformation, the system can brighten overall image while preventing noise emphasis in specific flat portions where it would cause scattered black dots.
Data Source
AI summary
An image processing unit of a processor for an endoscope includes: an emphasis processing calculation unit performing nonlinear gradation conversion for a pixel value of a pixel of interest, using each pixel of a captured image as the pixel of interest; and a preprocessing unit setting a reference upper limit characteristic line and a reference lower limit characteristic line in order to adjust an output pixel value after the nonlinear gradation conversion and calculating a degree of variation in pixel values in a partial setting region around the pixel of interest. The emphasis processing calculation unit calculates an output ratio of the output pixel value to a maximum pixel value that can be taken by the captured image, an adjustment upper limit value and an adjustment lower limit value, and an emphasis-processed pixel value, using the adjustment upper limit value, the adjustment lower limit value, and the output ratio.