Enhancement Infrared-Visible Image Fusion Using the Integration of Stationary Wavelet Transform and Fuzzy Histogram Equalization
Image fusion is the process of merging two or more images to obtain complementary features from source images. Imaging techniques in real-world applications provide images with a different texture than the other, where visible images provide spatial information while infrared images provide spectra...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
middle technical university
2022-12-01
|
| Series: | Journal of Techniques |
| Subjects: | |
| Online Access: | https://journal.mtu.edu.iq/index.php/MTU/article/view/700 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Image fusion is the process of merging two or more images to obtain complementary features from source images. Imaging techniques in real-world applications provide images with a different texture than the other, where visible images provide spatial information while infrared images provide spectral information. Hence the importance of image fusion, which aims to combine spatial and spectral information in one image. Wavelet transform is a method used in the process of image fusion as feature extraction, and images are decomposed into a series of low and high-frequency subbands. Wavelet transform provides images with good representation and is a multi-resolution analysis. However, the resulting image after the wavelet-based fusion process has low-quality information which is blurry. In addition, infrared images by their nature suffer from blur. In this paper, a novel image fusion method has been proposed to enhance visible-infrared image fusion using the integration of stationary wavelet transform and fuzzy histogram equalization. Firstly, input the images. Secondly, preprocessing the images. Thirdly, stationary wavelet transform has been used for decomposing the images in 2 levels. Fourthly, Averaging fusion rule is used for fusing the approximation coefficients. Finally, fuzzy histogram equalization is used in reconstructing the level 2 process to obtain the final enhanced image. The performance of the proposed method is evaluated by using seven metrics that proved the superiority of the proposed method compared to the standard methods.
|
|---|---|
| ISSN: | 1818-653X 2708-8383 |