Parallel optimization method of low-illumination image enhancement based on CUDA

An image enhancement and optimization method technology, applied in image enhancement, image data processing, instruments and other directions, can solve the problems of inability to meet real-time processing requirements, long running time, and large amount of calculation of image enhancement algorithms, and achieves real-time processing effect, The effect of reducing data transmission and reducing algorithm time-consuming
CN104881848AInactive Publication Date: 2015-09-02XIDIAN UNIV
6 Cites 13 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
XIDIAN UNIV
Publication Date
2015-09-02
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a parallel optimization method of low-illumination image enhancement based on CUDA. By adopting a CPU / GPU heterogeneous mode, low-illumination image enhancement algorithms are all performed on GPU, and input data and output data are copied between CPU and GPU. Three kernels are used on GPU; each kernel is provided with threads, the number of which is the same as that of image pixels; and the three kernels are responsible for image inversion and estimation of dark channel prior, estimation of overall atmospheric light, and calculation of transmissivity and haze-removal model and inversion operation of the image, respectively. In the low-illumination image enhancement algorithm based on dark channel prior haze-removal technology, calculation, not suitable to perform on GPU, of overall atmospheric light is improved; the atmospheric light value is estimated by using the brightness of dark channel prior and the image; and data relevance is reduced. According to the invention, a visual effect of a night image is improved, and a real-time treatment effect is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the field of parallel computing and the technical field of image processing algorithms, in particular to a CUDA-based parallel optimization method for low-illuminance image enhancement. Background technique

[0002] In order to improve the public safety of the society, video surveillance systems have been installed in many public places to understand the situation through monitoring images. However, when the light is too weak at night and the visibility is very low, it is difficult for the human eye to observe the surrounding environment. The images collected by image acquisition equipment at night are often dark, which reduces the visibility of the surrounding scenery and cannot meet some practical purposes. . Therefore, it is very meaningful in all aspects of life to use low-light image enhancement algorithms to process images and convert night images into images in which scene details can be seen during the day. Howeve...

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