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

Pulse-coupled image fusion method based on multi-channel mechanism

A technology of pulse coupling and image fusion, which is applied in the field of image processing, can solve problems such as adverse effects of fusion results, differences in fusion results, and lack of globality, and achieve the effects of overcoming differences in fusion results, increasing computing speed, and improving fusion effects

Inactive Publication Date: 2019-02-26
QINGDAO UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] (1) It is difficult to determine the order in which multiple images are fused, and different fusion sequences will lead to differences in the final fusion results
Experimental studies have found that different fusion sequences of multiple images will have a greater impact on the fusion results
In particular, the selection of the first few images is very important. If the result of fusion in the initial stage is not ideal, it will have a great adverse effect on the final fusion result.
[0005] (2) Due to the need to repeatedly call the dual-channel model, it will result in a large time cost
[0006] (3) During the fusion process of two images, the link weight information is determined according to the current situation of the two images. Due to the lack of globality, it cannot take into account the situation of all images, which will cause some valuable information to be lost. loss

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
  • Pulse-coupled image fusion method based on multi-channel mechanism
  • Pulse-coupled image fusion method based on multi-channel mechanism
  • Pulse-coupled image fusion method based on multi-channel mechanism

Examples

Experimental program
Comparison scheme
Effect test

experiment example 1

[0082] Experimental example 1: Comparative experiment with dual-channel pulse-coupled neural network image fusion method

[0083] In this experiment, multi-focus images are selected, and the related performance analysis is carried out by comparing with the dual-channel pulse-coupled neural network image fusion method.

[0084] Among them, the process of using the dual-channel model to realize the 4-channel fusion is as follows: arrange the four original images in different orders. The fusion result is sequentially fused with the third original image to be fused and the fourth original image to be fused using the same method to obtain the final fusion result. There are 12 situations in the four original images due to different arrangement orders. In the experiment, the fusion effects of these 12 situations were compared and analyzed in detail. In the experiment, the 4 original images, standard images, and dual channels are respectively 1-2-3-4, 1-2-4-3, 1-3-4-2, 1-3-2-4, 1-4- ...

experiment example 2

[0094] Experimental example 2: Comparative experiment with other pulse-coupled and non-pulse-coupled neural network image fusion methods

[0095] In order to further verify the effectiveness of the novel multi-channel pulse-coupled image fusion method in synchronous fusion of multiple images, this experiment example compares the method of the present invention with other methods. The specific comparison methods are: non-pulse-coupled neural network image fusion methods NSCT, LP, DWT, CP, MP, RP, SIDWT and pulse-coupled image fusion methods Wang-1, PCNN-1, PCNN-2, Fu-1. In the NSCT method, the high-frequency part is selected to be the largest, and the low-frequency part is selected to take the average fusion rule; when the non-impulse coupling method needs wavelet decomposition, the number of decomposition layers is set to 2, the high-frequency part adopts the fusion strategy of selecting the largest, and the low-frequency part adopts the selection of the second A fusion strate...

experiment example 3

[0101] Experimental example 3: N channel experiment (N>4)

[0102] In order to further verify the effectiveness of the method of the present invention, a large number of experimental studies have also been carried out when the number of channels exceeds 4, but considering factors such as experimental conditions and image source limitations, relevant experiments are only carried out on channels 5, 6 and 7 .

[0103]

[0104] Experimental results such as Figure 7As shown in Table 8, from the experimental results, the multi-channel model fusion method proposed by the present invention can also obtain a better fusion effect when the number of channels is more than 4. However, as the number of channels increases, the mean value of the fused image will continue to increase, which is related to the link weight calculation method used in this paper. At the same time, due to the increase in the number of channels that need to calculate the fusion information, the calculation amou...

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 relates to the technical field of image processing, and provides a pulse coupling image fusion method based on a multi-channel mechanism, comprising the following steps: S1, establishinga multi-channel pulse coupling neural network model structure, normalizing input image matrices of each channel, and inputting the input image matrices into the multi-channel pulse coupling neural network model structure; 2, calculating that value of each channel link weight beta ij alpha, and inputting the calculate link weight into a multi-channel impulse coupling neural network model structure; 3, carrying out fusion calculation in that fusion pool to obtain an internal output amount Uij [m]; 4, carrying out iterative calculation to generate a pulse ignition matrix Yij [m]; S5, taking thepulse ignition matrix Yij [m] as the gray matrix of the fusion image, generating a fusion image and outputting the fusion result. The invention can improve the fusion effect and efficiency, and can bewidely applied in the field of image fusion.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a pulse coupling image fusion method based on a multi-channel mechanism. Background technique [0002] The image fusion method based on the pulse-coupled neural network model has been effectively applied in many aspects. However, the current research is limited to the dual-channel pulse-coupled neural network image fusion model. With the continuous development of applications, the above methods still face some common limitations. With the continuous advancement of science and technology and the development of society, it is necessary to fuse multiple images into a clearer image that contains more information in many fields. At present, the specific method to solve this problem in the field of pulse-coupled neural network model is based on the dual-channel model. First, the two images are fused, and then the fused image is fused with the third, fourth... images in turn, ...

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): G06T5/50
CPCG06T5/50G06T2207/20084G06T2207/20221
Inventor 李奕赵俊莉吕智涵潘振宽
Owner QINGDAO UNIV
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com