Paper Info
Title
A novel multi-source image fusion method for pig-body multi-feature detection in NSCT domain
Abstract
The multi-source image fusion has been a hot topic during the recent years because of its higher detection rate. To improve the accuracy of pig-body multi-feature detection, a multi-source image fusion method was adopted in this field. However, the traditional multi-source image fusion methods could not obtain better contrast and more details of the fused image. To better detect shape and temperature feature of pig-body, a novel infrared and visible image fusion method was proposed in non-subsampled contourlet transform (NSCT) domain and named NSCT-GF-IAG. Through this technique, the visible and infrared images were first decomposed into a series of multi-scale and multi-directional sub-bands using NSCT. Then, to better represent the fine-scale of texture information and coarse-scale detail information, Gabor filter with even-symmetry and improved average gradient (IAG) were employed to fuse low-frequency and high-frequency sub-bands, respectively. Next, the fused coefficients were reconstructed into a final fusion image by inverse NSCT. Finally, the shape feature of pig-body was obtained by automatic threshold segmentation and optimized by morphological processing. Moreover, the highest temperature was extracted based on shape segmentation of pig-body. Experimental results showed that the proposed fusion method for detecting multi-feature was capable of achieving 2.175-5.129% higher average segmentation rate than the prevailing conventional methods. Besides this, the proposed method also improved efficiency in terms of time consumption.
Year
DOI
Venue
2020
10.1007/s11042-020-09044-9
MULTIMEDIA TOOLS AND APPLICATIONS
Keywords
DocType
Volume
Nonsubsampled contourlet transform,Gabor filter,Improved average gradient,Pig-body shape segmentation,Pig-body temperature detection
Journal
79.0
Issue
ISSN
Citations 
35-36
1380-7501
0
PageRank 
References 
Authors
0.34
0
4
Name
Order
Citations
PageRank
Zhen Zhong121.43
Wanlin Gao267.58
Abdul Mateen Khattak300.34
Minjuan Wang430441.52