SMOTE and its variants have been widely employed to complement minority classes and suppress the effect of class imbalance [55–65]. For example, Fan et al. [66] and Zhou et al. [67] combined SMOTE technique wit
To this first component of our loss function, we add two components that aim to downsize the overlapping region between the densities of the anomaly scores of both types of observations. In order to have control of the latter, we consider AUCDensity𝑢AUCDensityu and AUCDentsity𝑐AUCDentsit...
Process Monitoring Using Domain-Adversarial Probabilistic Principal Component Analysis: A Transfer Learning Framework. IEEE Trans. Ind. Inform. 2023, 19, 1436–1444. [Google Scholar] [CrossRef] Wu, Y.; Liu, X.; Wang, Y.; Li, Q.; Guo, Z.; Jiang, Y. Improved deep PCA and Kullback–...
For example, Ref. [20] proposed an improved KPCA method (Multi-way Kernel Principal Component Analysis, MKPCA), incorporating the concept of popular learning into the objective function of KPCA to preserve the local structure. This method ensures that the resulting feature space not only preserves...
Keywords: image fusion; principal component analysis network; lightweight deep learning network; image pyramid; infrared image1. Introduction An infrared (IR) sensor reflects the temperature or thermal radiation differences in a scene and captures thermal radiation objects in the dark or in smoke. How...