Skip to content

The original implementation of the paper "A Medical Image Fusion Method based on MDLatLRRv2"

Notifications You must be signed in to change notification settings

songxujay/MDLatLRRv2

Repository files navigation

D2-LRR: A Dual-Decomposed MDLatLRR Approach for Medical Image Fusion

Accepted by MVIPIT 2023

arxiv link : arxiv

main.m----------------------------主函数

fusion_all_kinds.m-----------------根据融合图像类型,对应进行融合

ConvertRGBtoYUV.m-------------将彩色图像(RGB)转换为YUV空间

ConvertYUVtoRGB.m-------------将YUV转换为RGB空间

fusion_base_parts.m--------------融合低频特征

vector_fusion_method.m---------返回融合结果

vector_fusion_detail_parts.m-----融合细节特征

vector_deep_latent.m-------------返回图像的低频、高频矩阵

vector_decomposition.m---------求取图像的低频、高频矩阵

test_images-----------------------测试数据文件夹

models----------------------------存储低秩投影矩阵文件夹

learning_projection_matrix--------学习低秩投影矩阵程序

fused_p_s--------------------------存储融合结果文件夹,主函数运行时,自动生成

About

The original implementation of the paper "A Medical Image Fusion Method based on MDLatLRRv2"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages