Offshore medium-long wave infrared image fusion method
A long-wave infrared and image fusion technology, applied in the field of image processing, can solve problems such as Gibbs phenomenon, blurred image edges, blurred edges, etc.
Pending Publication Date: 2021-04-06
BEIJING UNIV OF TECH
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] Currently commonly used image fusion methods are based on discrete wavelets, curvelet transform, and non-subsampled contourlets. The low-frequency fusion strategy is usually weighted average, and the high-frequency fusion strategy is based on the proportion of neighborhood coefficients. The image needs to be down-sampled during the wave transform process, resulting in the Gibbs phenomenon in the fused image, which blurs the edge of the image and reduces the image quality
However, the weighted average low-frequency fusion strategy does not fully take into account the brighter overall grayscale of long waves, and blurs the edges, resulting in the loss of fusion details. The high-frequency coefficient fusion strategy based on the proportion of neighborhood coefficients does not make full use of it. The respective edge details in the medium and long wave images lead to unsatisfactory fusion results. Therefore, a method based on NSCT decomposition and using full variational model and significance analysis to fuse high and low frequency coefficients is proposed. Both have improved the fusion effect
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0019] exist figure 2 China is the flow chart of overseas medium and long wave infrared image fusion method, including image decomposition based on NSCT, low frequency coefficient fusion based on total variation model, high frequency coefficient fusion based on saliency analysis and NSCT inverse transform image reconstruction.
[0020] Step S1, such as figure 1 Shown is a schematic diagram of the principle of image decomposition based on NSCT. In the specific implementation, the NSP transform in NSCT uses the 'pyrexc' filter, the NSDFB uses the 'vk' filter, the number of decomposition layers is set to 2, the decomposition direction is set to 8, medium and long wave Infrared original image such as image 3 As shown, the high-frequency sub-band coefficients and the low-frequency sub-band coefficients are obtained by the first-level decomposition, and the low-frequency coefficients are obtained by the first-level decomposition as the input of the next-level decomposition. with...
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
Login to View More Abstract
The invention discloses an offshore medium-long wave infrared image fusion method, which comprises the following steps of: decomposing a medium-long wave infrared image by utilizing NSCT (Non-subsampled Contourlet Transform) to obtain corresponding high-frequency and low-frequency components, fusing the low-frequency components of the medium-long wave infrared image through a total variation model, and calculating a weighting matrix through saliency analysis; and completing fusion of high-frequency components of the medium-long wave infrared image by using the matrix, and finally performing NSCT inverse transformation by combining high-frequency and low-frequency components obtained by fusion to obtain a final fused image. By the adoption of the fusion method, the overall brightness of the long-wave infrared medium image is reserved in a fusion result, edge details in the medium-wave infrared image are fused into a final result, meanwhile, interference of sea waves and a cloud layer is smoothed, the direction significance of a target is improved, and subsequent target detection is facilitated.
Description
technical field [0001] The invention belongs to the field of image processing, and in particular relates to a method for fusion of marine medium and long wave infrared images. Background technique [0002] At present, the commonly used image fusion methods are based on discrete wavelet, curvelet transform and non-subsampled contourlet. The low-frequency fusion strategy is usually weighted average, and the high-frequency fusion strategy is based on the proportion of neighborhood coefficients. In the process of wave transformation, the image needs to be down-sampled, which leads to the Gibbs phenomenon in the fused image, which blurs the edge of the image and reduces the image quality. However, the weighted average low-frequency fusion strategy does not fully take into account the brighter overall gray level of long-wavelength wavelengths, and blurs the edges, resulting in a loss of fusion details. The high-frequency coefficient fusion strategy based on the proportion of neigh...
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
Login to View More Patent Type & AuthorityApplications(China)
IPC IPC(8): G06T5/50
CPCG06T5/50G06T2207/10048G06T2207/20221G06T2207/20024
Inventor郑鑫陈良栋吴强康迪和薇代慧峰
OwnerBEIJING UNIV OF TECH