Abstract:An image fusion method based on fuzzy regional characteristics is proposed in this paper. After the multi-resolution decomposition of an image, k-mean clustering is firstly done for the low frequency components of the each layer to decompose the low frequency image into important region, sub important region and background region. Then, all areas of the image are fuzzificated and fusion strategies are determined according to their fuzzy membership degrees. Finally, fusion result is obtained by the reconstructi… Show more
“…According to the theoretical basis mentioned in Section 2.1, the cascade encoder is used to extract redundant and complementary features from infrared and visible images, and transforms these features into different spaces separately. The extracted redundant features and complementary features are fused with different fusion strategies softly directed by the fuzzy regional characteristics (FRC) method [20,21]. The FRCbased feature-weighted fusion strategy is used to fuse the redundancy feature, while the FRC-based feature choose-max fusion strategy is adopted to integrate the complementary feature.…”
Section: The Proposed Image Fusion Frameworkmentioning
confidence: 99%
“…Significantly, the region containing the target information in the redundant information can be effectively fused. Hence, it is suitable for complementary features and redundant features in the hidden layer to use the fusion rule that is designed based on the idea of the FRC method [20,21].…”
Section: Frc-based Fusion Strategymentioning
confidence: 99%
“…The joint cascade encoder is used to extract redundant and complementary feature maps from source images, enabling the feature maps to contain more source images information. (2) In the fusion layer, multiscale salient feature maps are merged by manually designed fusion strategies based on fuzzy region characteristics (FRC) [20,21]. The fusion strategy under the fuzzy region rule is adopted to protect regional consistency of the fused image.…”
To facilitate the extraction of source image information, and preserve the consistency of heterogeneous regional features, a multiscale image fusion method based on a joint cascaded convolutional coding network (JCa2Co) is proposed. The JCa2Co network can retain vast quantities of information from source images in a multiscale perspective. The approach includes an encoder, a fusion layer and decoder. In the fusion layer, the Fuzzy Regional Characteristics (FRC) scheme is considered, and the multiscale feature maps are extracted from image subregions to ensure regional image consistency. Firstly, a joint cascaded encoder is used to extract multiscale features of the source image, in which the output of each layer is connected to every other layer. The fusion layer based on FRC is then performed to fuse each scale feature. Finally, the fused image is reconstructed by the decoder. In addition, to verify the regional consistency of the fused image, a regional consistency measure is proposed. Experiments are performed on the TNO Image Fusion Database. The experimental results exhibit that the proposed JCa2Co method has better comprehensive performance than the eight state-of-the-art fusion methods. Moreover, it can effectively integrate meaningful information in infrared and visible images and has excellent performance in objective evaluation and visual quality, which is beneficial to target recognition and tracking.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
“…According to the theoretical basis mentioned in Section 2.1, the cascade encoder is used to extract redundant and complementary features from infrared and visible images, and transforms these features into different spaces separately. The extracted redundant features and complementary features are fused with different fusion strategies softly directed by the fuzzy regional characteristics (FRC) method [20,21]. The FRCbased feature-weighted fusion strategy is used to fuse the redundancy feature, while the FRC-based feature choose-max fusion strategy is adopted to integrate the complementary feature.…”
Section: The Proposed Image Fusion Frameworkmentioning
confidence: 99%
“…Significantly, the region containing the target information in the redundant information can be effectively fused. Hence, it is suitable for complementary features and redundant features in the hidden layer to use the fusion rule that is designed based on the idea of the FRC method [20,21].…”
Section: Frc-based Fusion Strategymentioning
confidence: 99%
“…The joint cascade encoder is used to extract redundant and complementary feature maps from source images, enabling the feature maps to contain more source images information. (2) In the fusion layer, multiscale salient feature maps are merged by manually designed fusion strategies based on fuzzy region characteristics (FRC) [20,21]. The fusion strategy under the fuzzy region rule is adopted to protect regional consistency of the fused image.…”
To facilitate the extraction of source image information, and preserve the consistency of heterogeneous regional features, a multiscale image fusion method based on a joint cascaded convolutional coding network (JCa2Co) is proposed. The JCa2Co network can retain vast quantities of information from source images in a multiscale perspective. The approach includes an encoder, a fusion layer and decoder. In the fusion layer, the Fuzzy Regional Characteristics (FRC) scheme is considered, and the multiscale feature maps are extracted from image subregions to ensure regional image consistency. Firstly, a joint cascaded encoder is used to extract multiscale features of the source image, in which the output of each layer is connected to every other layer. The fusion layer based on FRC is then performed to fuse each scale feature. Finally, the fused image is reconstructed by the decoder. In addition, to verify the regional consistency of the fused image, a regional consistency measure is proposed. Experiments are performed on the TNO Image Fusion Database. The experimental results exhibit that the proposed JCa2Co method has better comprehensive performance than the eight state-of-the-art fusion methods. Moreover, it can effectively integrate meaningful information in infrared and visible images and has excellent performance in objective evaluation and visual quality, which is beneficial to target recognition and tracking.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.