Treffer: Intelligent classification techniques fuzzy CNN use face detection using deep learning.

Since many practical applications have become auto mutable, face recognition has become an increasingly important problem in recent years. The ability to recognize a candidate's face is crucial in the test system. People using counterfeit keys at airports may be identified with the use of FR systems. In addition to its use in the military services, homeland security, FR-based healthcare information systems, banking, and reservation systems all benefit from it. Finding ways to apply Hyper Spectral Imaging (HSI) techniques to face datasets is crucial when looking to increase the datasets dimensional. The use of biometric features, such as spook functions, with hyper spectral imaging methods has resulted to increased accuracy in face recognition. However, the effectiveness of 3-D image-based techniques decreases as the number of faces to be assessed increases due to the reduction in the inter-object space in the facial recognition domain. Due to the large size of the features, an active spooky imaging method is required to improve the problem's efficacy. Because of this, combining HSI with AI methods like Machine Learning (ML) is essential for reliable face recognition. In this paper, we propose three novel formulae that combine deep semantic network methods with conventional knowledge approaches. In this work, deep learning techniques are used to complete the task of face recognition as these algorithms automatically complete the processes of attribute extraction, attribute selection, and classification. Better efficiency with regard to category accuracy may also be achieved by combining the throng intelligence approaches with deep learning algorithms. Therefore, this thesis suggests new formulas by integrating flock intelligence formulas—such as firefly, Dragonfly, and grey-wolf optimization—with deep learning formulas like Convolution Neural Network (CNN), Recursive Semantic Network (RNN), and Vision Transformer (ViT). Here, we suggest using a combination of CNN and firefly, RNN and dragonfly, or an ambiguous CNN with uncertainty bounds. [ABSTRACT FROM AUTHOR]

Copyright of AIP Conference Proceedings is the property of American Institute of Physics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)