Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Full-scale DR image enhancement processing method optimized for GPU

A processing method and image enhancement technology, applied in the field of image processing, can solve the problems of inability to meet the requirements of abnormal exposure image processing, abnormally prominent lung texture, bone texture, sacrificing the tolerance of processing methods, etc., to ensure accuracy, maintain Tolerance, the effect of improving calculation speed

Pending Publication Date: 2018-07-03
河北天地智慧医疗设备股份有限公司
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although these processing methods can meet the processing effect of ordinary digital images, for medical DR images with extremely high tissue gray level complexity, the processing effect often leads to excessively prominent detail textures, resulting in abnormally prominent lung textures and bone textures, which is easy to Misleading doctors to produce misdiagnosis
If you simply increase the grayscale / contrast limitation, the tolerance of the processing method will often be sacrificed, and the image processing requirements for abnormal exposure (high or low overall grayscale) will not be met.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Full-scale DR image enhancement processing method optimized for GPU
  • Full-scale DR image enhancement processing method optimized for GPU
  • Full-scale DR image enhancement processing method optimized for GPU

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach

[0032] refer to Figure 1-4 , a specific embodiment of the present invention comprises the following steps:

[0033] A. Read the preset full-scale median filter adaptive curve, full-scale contrast adjustment curve and full-scale fast Gaussian filter adjustment curve; the original image data is directly imported into the video memory, and all processing operations use GPU;

[0034] B. Perform noise reduction processing on the whole image; obtain its gray value G according to the coordinates (x, y) of each pixel xy , the median filter range z of the pixel is calculated from the preset full-scale median filter adaptive curve, and the median value G of all pixels in the (x-z, y-z)-(x+z, y+z) rectangular range is calculated m as the gray value of the pixel;

[0035] C. Carry out histogram prescriptive processing on the whole image;

[0036] D. Divide the image into 16x16 blocks;

[0037] E. Process each block separately; for the block whose coordinates are (m, n), calculate its...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a full-scale DR image enhancement processing method optimized for GPU (Graphics Processing Unit), comprising the following steps: A, directly importing original image data to avideo memory, and using a GPU in all processing operations; B, de-noising the whole image; C, carrying out histogram normalization on the whole image; D, dividing the image into 16*16 blocks; E, processing each block; F, splicing the blocks into an image, and eliminating the boundary effect of each block; G, filtering and sharpening the whole image; H, adjusting the overall brightness and contrast of the whole image; and I, directly outputting the final image to a display, and exporting the data to a memory. The method can overcome the shortage of the prior art, can achieve a good processingeffect for images with too high or low exposure dose as well as images with high tissue density complexity, and greatly reduces the time for image output.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a GPU-optimized full-scale DR image enhancement processing method. Background technique [0002] With the development of digital imaging medical equipment, the digital image processing technology collected by medical equipment directly affects the quality of the final image and the accuracy of the final diagnosis result. In order to improve the recognizability of key information in images, various image enhancement methods are widely used in the processing of digital medical images. [0003] Common image processing methods include noise reduction, histogram transformation, filter enhancement, etc. Usually, the medical image processing process includes the above-mentioned various processing modules, which are processed step by step. Due to the high resolution of DR images (usually about 3072x3072), conventional image processing methods use CPU calculations, which take a ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T5/00G06T5/40
CPCG06T5/40G06T5/92G06T5/73G06T5/70
Inventor 蒲海杨子寒耿亚洲刘超
Owner 河北天地智慧医疗设备股份有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products