An infrared and visible image fusion method based on improved DenseNet and mRMR-ZCA
K Ren, D Zhang, M Wan, X Miao, G Gu… - Infrared Physics & …, 2021 - Elsevier
Extracting sufficient features with limited available information is essential for image fusion.
However, most of the deep learning algorithms cannot fully extract the features. This paper
proposes a novel infrared and visible images fusion method based on improved DenseNet,
Max-Relevance and Min-Redundancy (mRMR) and zero-phase component analysis (ZCA).
Firstly, the convolutional layer and Dense block are utilized to extract the dense features of
the source images. Secondly, the mRMR operation is used to filter out the feature set that …
However, most of the deep learning algorithms cannot fully extract the features. This paper
proposes a novel infrared and visible images fusion method based on improved DenseNet,
Max-Relevance and Min-Redundancy (mRMR) and zero-phase component analysis (ZCA).
Firstly, the convolutional layer and Dense block are utilized to extract the dense features of
the source images. Secondly, the mRMR operation is used to filter out the feature set that …