This paper investigates the ability of the depolarization degree, derived from the characteristic polarization states at the resonant frequency set, to identify corner or swept, i.e. dihedral, changes in same-class ta...This paper investigates the ability of the depolarization degree, derived from the characteristic polarization states at the resonant frequency set, to identify corner or swept, i.e. dihedral, changes in same-class targets by a metallic wire example. A well-estimated depolarization degree requires a robust extraction of the fundamental target resonance set in two orthogonal sets of fully co-polarized and cross-polarized polarization channels, then finding the null polarization states using the Lagrangian method. Such depolarization degree per resonance mode has the potential to form a robust feature set because it is relatively less sensitive to onset ambiguity, invariant to rotation, and could create a compact, recognizable, and separable distribution in the proposed feature space. The study was limited to two targets with two swept changes of fifteen degrees within normal incidence;under a supervised learning approach, the results showed that the identification rate converging to upper-bound (100%) for a signal-to-noise ratio above 20 dB and lower-bound around (50%) below −10 dB.展开更多
Unsupervised methods based on density representation have shown their abilities in anomaly detection,but detection performance still needs to be improved.Specifically,approaches using normalizing flows can accurately ...Unsupervised methods based on density representation have shown their abilities in anomaly detection,but detection performance still needs to be improved.Specifically,approaches using normalizing flows can accurately evaluate sample distributions,mapping normal features to the normal distribution and anomalous features outside it.Consequently,this paper proposes a Normalizing Flow-based Bidirectional Mapping Residual Network(NF-BMR).It utilizes pre-trained Convolutional Neural Networks(CNN)and normalizing flows to construct discriminative source and target domain feature spaces.Additionally,to better learn feature information in both domain spaces,we propose the Bidirectional Mapping Residual Network(BMR),which maps sample features to these two spaces for anomaly detection.The two detection spaces effectively complement each other’s deficiencies and provide a comprehensive feature evaluation from two perspectives,which leads to the improvement of detection performance.Comparative experimental results on the MVTec AD and DAGM datasets against the Bidirectional Pre-trained Feature Mapping Network(B-PFM)and other state-of-the-art methods demonstrate that the proposed approach achieves superior performance.On the MVTec AD dataset,NF-BMR achieves an average AUROC of 98.7%for all 15 categories.Especially,it achieves 100%optimal detection performance in five categories.On the DAGM dataset,the average AUROC across ten categories is 98.7%,which is very close to supervised methods.展开更多
文摘This paper investigates the ability of the depolarization degree, derived from the characteristic polarization states at the resonant frequency set, to identify corner or swept, i.e. dihedral, changes in same-class targets by a metallic wire example. A well-estimated depolarization degree requires a robust extraction of the fundamental target resonance set in two orthogonal sets of fully co-polarized and cross-polarized polarization channels, then finding the null polarization states using the Lagrangian method. Such depolarization degree per resonance mode has the potential to form a robust feature set because it is relatively less sensitive to onset ambiguity, invariant to rotation, and could create a compact, recognizable, and separable distribution in the proposed feature space. The study was limited to two targets with two swept changes of fifteen degrees within normal incidence;under a supervised learning approach, the results showed that the identification rate converging to upper-bound (100%) for a signal-to-noise ratio above 20 dB and lower-bound around (50%) below −10 dB.
基金This work was supported in part by the National Key R&D Program of China 2021YFE0110500in part by the National Natural Science Foundation of China under Grant 62062021in part by the Guiyang Scientific Plan Project[2023]48-11.
文摘Unsupervised methods based on density representation have shown their abilities in anomaly detection,but detection performance still needs to be improved.Specifically,approaches using normalizing flows can accurately evaluate sample distributions,mapping normal features to the normal distribution and anomalous features outside it.Consequently,this paper proposes a Normalizing Flow-based Bidirectional Mapping Residual Network(NF-BMR).It utilizes pre-trained Convolutional Neural Networks(CNN)and normalizing flows to construct discriminative source and target domain feature spaces.Additionally,to better learn feature information in both domain spaces,we propose the Bidirectional Mapping Residual Network(BMR),which maps sample features to these two spaces for anomaly detection.The two detection spaces effectively complement each other’s deficiencies and provide a comprehensive feature evaluation from two perspectives,which leads to the improvement of detection performance.Comparative experimental results on the MVTec AD and DAGM datasets against the Bidirectional Pre-trained Feature Mapping Network(B-PFM)and other state-of-the-art methods demonstrate that the proposed approach achieves superior performance.On the MVTec AD dataset,NF-BMR achieves an average AUROC of 98.7%for all 15 categories.Especially,it achieves 100%optimal detection performance in five categories.On the DAGM dataset,the average AUROC across ten categories is 98.7%,which is very close to supervised methods.
