A Research of Gas Open-Set Identification Based on Data Augmentation Algorithm

Significant progress has been made in convolutional neural networks (CNN) based gas recognition. However, existing electronic nose (e-Nose) algorithms all use the closed-set assumption that test and training samples are same label space can only detect objects of known classes. realistic scenarios, collecting data for every possible would waste much resource. Open-set identification aims to actively reject from unknown classes by reducing intra-class spacing and, thus, not misclassifying them as In this study, we propose a preprocessing method enhance performance recognition augmenting eigenvalues each gas. We then implement open-set task gases using an model. These methods contribute improved accuracy provide effective means handling class samples. Experimental results show our approach identify well while maintaining available