Self-adaptive coding method based on low-resolution prior spectral image region segmentation
An adaptive coding, low-resolution technology, applied in image coding, image analysis, image enhancement and other directions, can solve problems such as distortion
Active Publication Date: 2021-05-18
北京理工大学重庆创新中心
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0008] The purpose of the present invention is to provide an adaptive coding method based on low-resolution priori spectral image region segmentation to solve the problem that traditional compression coding methods usually require scene high-resolution priori information, and the current There are distortion problems when using the low-resolution prior information of the scene to generate the encoding matrix to reconstruct the hyperspectral image
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 1
[0049] An adaptive coding method based on low-resolution prior spectral image region segmentation, such as figure 1 , 2 shown, including:
[0050] A. During the imaging process of scene information by a spectral scan, the compressed spectral imaging system (technology) is used to load the snapshot encoding template to obtain the low-resolution scene information of the scene as the spatial spectral prior information.
[0051] Assume that the spatial spectral information of the original scene is F, the spatial dimension coordinates are (x, y), and the spectral dimension coordinates are λ i (Assume that the spatial spectral information of the scene contains the wavelength λ 1 ,λ 2 ,λ 3 ,...,λ i ,...,λ L The scene information at the L bands of ) is f(x,y,λ i ). Under the band selection of the ideal spectral filter, only the central wavelength is allowed to be λ i The scene information of each band (λ 1 ,λ 2 ,λ 3 ,...,λ i ,...,λ L ) scene information.
[0052] Design...
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 a self-adaptive coding method based on low-resolution prior spectral image region segmentation, which comprises the following steps of: loading a snapshot type coding template by using a compressed spectral imaging system, and acquiring low-resolution scene information of a scene; segmenting the low-resolution scene information by using a segmentation-combination method to obtain a segmentation result of the low-resolution scene information; under a low-resolution self-adaptive coding framework, generating a region-based threshold matrix by using the segmentation result, and generating a region-based self-adaptive coding matrix in combination with approximate estimation of an original spectral image; and performing compression observation on the scene by using the adaptive coding matrix, and reconstructing scene information of the scene from a compression observation result by using a reconstruction algorithm. According to the method, sufficient high-resolution prior information is not needed, the spatial spectrum correlation of low-resolution scene information is mined through the adopted segmentation method, self-adaptive coding is generated with the area as the unit, and compared with a traditional coding mode, the quality of a reconstructed spectrum image is improved.
Description
technical field [0001] The invention relates to the technical field of compressed spectral imaging, in particular to an adaptive encoding method based on low-resolution priori spectral image region segmentation. Background technique [0002] Compressed sensing spectral imaging technology is a technology that utilizes the principle of compressed sensing to realize compression coding through various technical means in the process of collecting spatial spectral information of the scene to obtain a small amount or low resolution information of the scene, and reconstruct a large number or high resolution information of the scene through the reconstruction process. technology for resolution information. In 2007, Baraniuk and others at Rice University first successfully used this technology in the imaging field to make a single pixel camera (Single Pixel Camera, SPC). Researchers from Duke University, University of Arizona, University of Delaware, and domestic research institution...
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 & Authority Applications(China)
IPC IPC(8): G06T9/00G06T7/11G06T7/136
CPCG06T9/00G06T7/11G06T7/136G06T2207/10036
Inventor 许廷发张宇寒秦庆旺王茜孙丽敏黄旭
Owner 北京理工大学重庆创新中心