Low-light image enhancement method based on infrared information

An image enhancement, infrared image technology, applied in the field of image processing, can solve problems such as inability to solve image missing areas

Pending Publication Date: 2020-03-17
ANHUI UNIVERSITY
View PDF4 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Regarding low-light enhancement tasks, there are not a few deep learning methods currently used, but these methods also take into account the introduction of infrared information, and thes

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
  • Low-light image enhancement method based on infrared information
  • Low-light image enhancement method based on infrared information
  • Low-light image enhancement method based on infrared information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0179] Image 6 Shown is the performance of the comparison algorithm and our algorithm on the outdoor test set. From the figure, it can be clearly shown that compared with other algorithms, our algorithm recovers more details in the case of local low light. And other algorithms don't work well in areas where the light is too low. Therefore, this also directly proves the necessity of introducing infrared information in the task of low-light image restoration

Embodiment 2

[0181] Such as Figure 7 As shown, the current low-light image enhancement algorithms have not considered the extremely low-light situation, and our algorithm, because of the introduction of infrared information, has solved the problem of partial missing of the obtained image under extremely low light. In order to better demonstrate this, we not only reduce the brightness and contrast of the test set, but also randomly select 2 to 3 positions on the image to generate a square missing block with a length of 50 to 70 pixels (let pixel value 0) to simulate extremely low light conditions. Since the comparison algorithm cannot restore the missing area very well, we will no longer show the image rendering of the comparison algorithm with the missing area.

Embodiment 3

[0183] Figure 8 shows our ablation experiments. Such as image 3 In the network with scene attention layer, we design a dual-branch structure: one branch is with a foreground attention layer, and its main function is to reconstruct the features of missing regions. The second branch has multiple layers of dilated convolutions, whose function is to increase the receptive field of the network. In order to better demonstrate the working condition of the foreground attention layer, we delete one of the second-stage networks when the first-stage network remains unchanged, so as to observe the changes in the output results.

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

In order to overcome the defects of the background technology, the invention provides an end-to-end two-stage neural network model to fuse low-light and infrared information to enhance an image so asto achieve the purpose of enhancing a low-light picture. Firstly, we find that local information loss is caused by low light of a low-light picture, and lost information cannot be supplemented by simply enhancing the low-light picture. Inspired by the problem, information in two states is mutually supplemented by considering and combining an infrared image under low light. In the whole model, in the first stage, brightness and contrast improvement is carried out on an area where image pixels are not zero. A foreground attention module is applied to the second part of the network to reconstructan image of the missing area, so that a complete bright image is finally obtained. The beneficial technical effects are that compared with other algorithms, the algorithm introduces the infrared information, and more details are recovered under the condition of local low light, and the enhanced image is more natural as a whole, and the algorithm can process a darker image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a low-light image enhancement method based on infrared information. Background technique [0002] High-quality images play a vital role in computer vision tasks such as object recognition and scene detection. But in reality, the captured image quality is often degraded due to weather and lighting effects. For example, when a picture is taken under low-light conditions, the contrast and brightness of the image are very low, which greatly increases the difficulty of subsequent advanced tasks, and also reduces the algorithm performance under visible light. figure 1 Three such examples are provided, and it is evident from the images that many details are hidden into the dark background. In order to reproduce these details, low-light image enhancement is required. [0003] A typical choice for low-light enhancement is the histogram equalization strategy. This method is to ba...

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
IPC IPC(8): G06T5/40G06N3/04G06N3/08
CPCG06T5/40G06N3/08G06N3/045
Inventor 汪粼波杨德云方贤勇
Owner ANHUI UNIVERSITY
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap