SDDTT Medical Video Processing for Near-Real-View Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video processors face challenges in producing high-quality images from videoscopes due to noise and varying pixel intensities, which are exacerbated by factors like moisture, occlusion, and conflicting effects of denoising and sharpening, leading to increased costs and limitations in improving image quality, especially for single-use devices.
Innovation Solution
A single denoising and edge detection trained network (SDDTT) processes images in real-time to generate near real-view images, reducing noise and enhancing sharpness, allowing lower-cost videoscopes to produce images approaching higher resolution quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If denoising processing is applied to images from videoscopes, then noise is reduced, but image sharpness and detail are lost
Solution Approach 1:
The image processing is segmented into distinct functional blocks: a denoising block that removes noise while preserving edges, and a sharpening block that enhances image sharpness. This segmentation allows each block to optimize its specific function without interfering with the other, resolving the contradiction between noise reduction and sharpness preservation
Solution Approach 2:
Denoising is performed as a preliminary action before sharpening. By removing noise first and then applying sharpening to the already denoised image, the system avoids amplifying noise during the sharpening process, thus achieving both noise reduction and sharpness enhancement
2Manufacturing precision
If higher resolution image sensors are used in videoscopes, then image quality is improved, but device cost increases
Solution Approach 1:
Instead of using expensive high-resolution image sensors, the system uses lower-resolution sensors combined with image processing algorithms that create enhanced quality images. The processing pipeline acts as a computational copy or surrogate for the physical high-resolution sensor, achieving similar visual quality at lower hardware cost
Solution Approach 2:
The system changes the parameters of image processing (denoising strength, sharpening intensity, gamma correction) to optimize image quality output from lower-resolution sensors. By adjusting these processing parameters, the system compensates for the lower native resolution, achieving high-quality images without requiring expensive high-resolution sensors
3Manufacturing precision
If advanced image processing techniques are implemented, then image quality is improved, but processing complexity and cost increase
Solution Approach 1:
The complex image processing is segmented into three manageable blocks: denoising, gamma correction, and sharpening. Each block performs a specific function with optimized parameters, making the overall complex process more manageable and easier to implement in hardware or software without requiring excessive computational resources
Solution Approach 2:
The system applies targeted processing actions only where needed: denoising is applied to reduce noise in specific image regions, gamma correction adjusts brightness characteristics, and sharpening enhances edges. This partial application of processing techniques achieves high image quality without the need for exhaustive processing of the entire image, reducing overall complexity
Data Source
AI summary
A method to generate near real-view images and an image processor configured to execute the method. The method includes, by an image processing circuit connected to a videoscope: processing a source image corresponding to an image captured by the videoscope with a single denoising and edge detection trained network (SDDTT), the SDDTT outputting, in a single pass, a denoise map and an edge map; denoising the source image with the noise map to produce a denoised image; gamma-correcting the denoised image to produce a gamma-corrected image; and sharpening the gamma-corrected denoised image with the edge map to produce the near real-view image.