Image real-time dehazing method based on graphics processor

A graphics processor and image technology, applied in the field of image processing, can solve the problems that GPU computing does not fully tap data reusability, it is difficult to achieve real-time processing, and cannot achieve fast processing, so as to improve data access speed, overcome the impact of interruption, The effect of reducing the amount of calculation

Active Publication Date: 2019-05-24
XIDIAN UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the parallelism of the whole implementation scheme is not enough, some processes still use the CPU for serial processing, the GPU calculation part does not fully exploit the reusability of data, and does not use high-speed memory such as texture storage and shared storage to access data The mode is optimized so that it can only process low-resolution video images in real time. When the image resolution is slightly higher, it cannot achieve fast processing. For example, for a video image with a resolution of 720×576, its processing speed is only 21 frame / second
Obviously, it is difficult to achieve real-time processing for high-definition video images such as 1080p

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
  • Image real-time dehazing method based on graphics processor
  • Image real-time dehazing method based on graphics processor
  • Image real-time dehazing method based on graphics processor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] Attached below figure 1 The present invention is further described.

[0036] Step 1, initialize the computer.

[0037] Connect the graphics processor GPU with the computer, the graphics processor GPU refers to the Tesla-C2075 graphics processor GPU hardware architecture;

[0038] The computer allocates shared memory, texture memory and global memory for the graphics processor GPU to store the data when the graphics processor GPU performs parallel computing. These data include the image data of the input fog image and the image data of the output clear image And the intermediate results output by each kernel function.

[0039] Step 2, input the fog map I and read the image data.

[0040]Input the fog image I into the computer memory and read its image data, convert the integer data into floating point data and store it in the global memory.

[0041] Step 3, calculate the dark channel map I of the input fog map I dark .

[0042] Using a single thread to calculate a ...

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 an image real-time defogging method based on a graphics processor, mainly solving the problem that in the prior art, a computer CPU (Central Processing Unit) has a slow processing speed, and is not suitable for performing real-time defogging on high-definition images. The method comprises the steps of: 1, initializing a computer, inputting a fogging image and reading image data; 2, calculating a dark channel image for inputting the fogging image; 3, calculating a airglow value according to the dark channel image; 4, calculating the initial transmissivity image for inputting the fogging image; 5, guiding the initial transmissivity image to be filtered to obtain a final transmissivity image; and 6, restoring a photographic field according to the final transmissivity image, and outputting a clear and defogged image. The method can guarantee better defogging effects, has a fast processing speed, meets the real-time defogging requirement of a 10180p image, and can be used for high-definition image real-time defogging processing.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a real-time image defogging method, which can be used in the fields of outdoor target recognition and tracking, road visual monitoring, satellite remote sensing monitoring and the like. Background technique [0002] The generation of fog image is mainly due to the scattering effect of turbid media such as fog and water vapor in the atmosphere, which seriously attenuates the light intensity received by the camera. Therefore, in order to ensure the clarity of the image and the recognizability of information, and to correct the color shift of the image due to atmospheric light, the image defogging algorithm is usually used to process the fog image. The image defogging algorithm based on the dark channel prior physically defoggs the image, and the effect is obvious and real and natural, so it is widely used in the field of image defogging. [0003] However, due to the...

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 Patents(China)
IPC IPC(8): G06T5/00
CPCG06T5/003G06T2207/10016G06T2207/10024G06T2207/20024
Inventor 吴宪云王茹李云松雷杰王柯俨刘凯
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products