Multi-Fake Evolutionary Generative Adversarial Networks for Imbalance Hyperspectral Image Classification
Tanmoy Dam, Nidhi Swami, Sreenatha G. Anavatti, Hussein A. Abbass
TL;DR
This work tackles imbalanced hyperspectral image classification by introducing Multi-Fake Evolutionary ADversarial Networks (MFEGAN), an end-to-end framework that embeds a classifier on the GAN discriminator and uses multiple generator objective losses selected via an evolutionary strategy. Extending ACGAN, MFEGAN deploys an evolving generator that yields multiple fake class outputs and a fitness-driven rule to select the best generator loss, balancing sample quality and diversity. Evaluations on Indian Pines and Kennedy Space Center datasets show MFEGAN consistently outperforms ACGAN, AGGAN, and oversampling baselines across OA, AA, and Kappa, with McNemar's test confirming statistical significance. The approach demonstrates robust improvements for spatial-spectral hyperspectral data and suggests further gains from combining multiple spatial-spectral generators in future work.
Abstract
This paper presents a novel multi-fake evolutionary generative adversarial network(MFEGAN) for handling imbalance hyperspectral image classification. It is an end-to-end approach in which different generative objective losses are considered in the generator network to improve the classification performance of the discriminator network. Thus, the same discriminator network has been used as a standard classifier by embedding the classifier network on top of the discriminating function. The effectiveness of the proposed method has been validated through two hyperspectral spatial-spectral data sets. The same generative and discriminator architectures have been utilized with two different GAN objectives for a fair performance comparison with the proposed method. It is observed from the experimental validations that the proposed method outperforms the state-of-the-art methods with better classification performance.