As image manipulation technology advances rapidly,the malicious use of image tampering has alarmingly escalated,posing a significant threat to social stability.In the realm of image tampering localization,accurately l...As image manipulation technology advances rapidly,the malicious use of image tampering has alarmingly escalated,posing a significant threat to social stability.In the realm of image tampering localization,accurately localizing limited samples,multiple types,and various sizes of regions remains a multitude of challenges.These issues impede the model’s universality and generalization capability and detrimentally affect its performance.To tackle these issues,we propose FL-MobileViT-an improved MobileViT model devised for image tampering localization.Our proposed model utilizes a dual-stream architecture that independently processes the RGB and noise domain,and captures richer traces of tampering through dual-stream integration.Meanwhile,the model incorporating the Focused Linear Attention mechanism within the lightweight network(MobileViT).This substitution significantly diminishes computational complexity and resolves homogeneity problems associated with traditional Transformer attention mechanisms,enhancing feature extraction diversity and improving the model’s localization performance.To comprehensively fuse the generated results from both feature extractors,we introduce the ASPP architecture for multi-scale feature fusion.This facilitates a more precise localization of tampered regions of various sizes.Furthermore,to bolster the model’s generalization ability,we adopt a contrastive learning method and devise a joint optimization training strategy that leverages fused features and captures the disparities in feature distribution in tampered images.This strategy enables the learning of contrastive loss at various stages of the feature extractor and employs it as an additional constraint condition in conjunction with cross-entropy loss.As a result,overfitting issues are effectively alleviated,and the differentiation between tampered and untampered regions is enhanced.Experimental evaluations on five benchmark datasets(IMD-20,CASIA,NIST-16,Columbia and Coverage)validate the effectiveness of our proposed model.The meticulously calibrated FL-MobileViT model consistently outperforms numerous existing general models regarding localization accuracy across diverse datasets,demonstrating superior adaptability.展开更多
We investigate the linear momentum density of light, which can be decomposed into spin and orbital parts, in the complex three-dimensional field distributions of tightly focused vortex segmented beams. The chosen angu...We investigate the linear momentum density of light, which can be decomposed into spin and orbital parts, in the complex three-dimensional field distributions of tightly focused vortex segmented beams. The chosen angular spectrum exhibits two spatially separated vortices of opposite charge and orthogonal circular polarization to generate phase vortices in a meridional plane of observation. In the vicinity of those vortices, regions of negative orbital linear momentum occur. Besides these phase vortices, the occurrence of transverse orbital angular momentum manifests in a vortex charge-dependent relative shift of the energy density and linear momentum density.展开更多
基金This study was funded by the Science and Technology Project in Xi’an(No.22GXFW0123)this work was supported by the Special Fund Construction Project of Key Disciplines in Ordinary Colleges and Universities in Shaanxi Province,the authors would like to thank the anonymous reviewers for their helpful comments and suggestions.
文摘As image manipulation technology advances rapidly,the malicious use of image tampering has alarmingly escalated,posing a significant threat to social stability.In the realm of image tampering localization,accurately localizing limited samples,multiple types,and various sizes of regions remains a multitude of challenges.These issues impede the model’s universality and generalization capability and detrimentally affect its performance.To tackle these issues,we propose FL-MobileViT-an improved MobileViT model devised for image tampering localization.Our proposed model utilizes a dual-stream architecture that independently processes the RGB and noise domain,and captures richer traces of tampering through dual-stream integration.Meanwhile,the model incorporating the Focused Linear Attention mechanism within the lightweight network(MobileViT).This substitution significantly diminishes computational complexity and resolves homogeneity problems associated with traditional Transformer attention mechanisms,enhancing feature extraction diversity and improving the model’s localization performance.To comprehensively fuse the generated results from both feature extractors,we introduce the ASPP architecture for multi-scale feature fusion.This facilitates a more precise localization of tampered regions of various sizes.Furthermore,to bolster the model’s generalization ability,we adopt a contrastive learning method and devise a joint optimization training strategy that leverages fused features and captures the disparities in feature distribution in tampered images.This strategy enables the learning of contrastive loss at various stages of the feature extractor and employs it as an additional constraint condition in conjunction with cross-entropy loss.As a result,overfitting issues are effectively alleviated,and the differentiation between tampered and untampered regions is enhanced.Experimental evaluations on five benchmark datasets(IMD-20,CASIA,NIST-16,Columbia and Coverage)validate the effectiveness of our proposed model.The meticulously calibrated FL-MobileViT model consistently outperforms numerous existing general models regarding localization accuracy across diverse datasets,demonstrating superior adaptability.
文摘We investigate the linear momentum density of light, which can be decomposed into spin and orbital parts, in the complex three-dimensional field distributions of tightly focused vortex segmented beams. The chosen angular spectrum exhibits two spatially separated vortices of opposite charge and orthogonal circular polarization to generate phase vortices in a meridional plane of observation. In the vicinity of those vortices, regions of negative orbital linear momentum occur. Besides these phase vortices, the occurrence of transverse orbital angular momentum manifests in a vortex charge-dependent relative shift of the energy density and linear momentum density.
