Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68080
Volume:
10 |
10 |
Oct 2026
Pages:
1-4
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science
St. Joseph's First Grade College, Jayalakshmipuram,Mysore
In recent years, biometric identification has gained prominence due to its potential for secure and reliable authentication. Among the various biometric modalities, ear biometrics stands out as a non-intrusive and stable method for identity verification and gender classification. This project explores the use of deep learning techniques to classify gender based on ear images, aiming to overcome the limitations of traditional methods like facial recognition and fingerprint scanning, which are often affected by occlusions, expression changes, and aging. The proposed system follows a structured pipeline that includes user registration, image acquisition, preprocessing, feature extraction, and classification. Preprocessing techniques such as grayscale conversion, resizing, noise reduction, and contrast enhancement are applied to improve image quality. For feature extraction and classification, the system employs VGG16, a pre-trained Convolutional Neural Network (CNN) known for its robust performance in image classification tasks. The model is trained and evaluated on a dataset of ear images, ensuring diversity in gender representation. The experimental results demonstrate that the proposed deep learning-based system achieves high accuracy in gender prediction, outperforming traditional handcrafted feature-based approaches. The stability of ear structure over time enhances the model’s reliability, making it suitable for applications in security systems, forensic investigations, and personalized user experiences. This project illustrates the potential of combining biometric science with deep learning to create fast, accurate, and non-intrusive gender classification systems.
KEYWORDS
This study successfully developed a gender classification system based on ear biometrics using deep learning techniques. By leveraging the VGG16 convolutional neural network for feature extraction and combining it with a customized classification layer, the system achieved a high accuracy of 94.2%, outperforming many traditional face-based classification methods. The use of ear images as the primary biometric modality proved to be highly effective due to their stability, uniqueness, and resistance to changes caused by facial expressions or occlusions. The project demonstrates that ear-based biometrics can be a viable and robust alternative to conventional facial recognition methods, especially in real-world conditions where facial data may be unreliable. The preprocessing pipeline, deep learning architecture, and performance metrics collectively validate the reliability and efficiency of the proposed system.
[1] [1] A. Saxena, P. Singh and S. Narayan Singh, "Gender and Age detection using Deep Learning," 2021 11th International Conference on Cloud Computing, Data Science & Engineering (Confluence), 2021, pp. 719-724, [2] [2] M. Vaibhav, R. Vanshaj, J. Vanshika, D. Anil Kumar, Compute Age and Gender of Person through ML, International Research Journal of Engineering and Technology (IRJET), Volume 07, Issue 06, 2020, Pages 732-737, ISSN 2395-0056 [3] [3] Agrawal, B., Dixit, M. (2020). Age Estimation and Gender Prediction Using Convolutional Neural Network. In: Pandit, M., Srivastava, L., Venkata Rao, R., Bansal, J. (eds) Intelligent Computing Applications for Sustainable Real-World Systems. ICSISCET 2019. Proceedings in Adaptation, Learning and Optimization, vol 13. Springer, Cham. [4] [4] Chao Yin and Xin Geng,” Facial Age Estimation by Conditional Probability Neural Network”, 2012 Springer Verlag Berlin Heidelberg [5] [5] https://www.kaggle.com/jangedoo/utkface-new [6] [6] https://www.tensorflow.org/api_docs/python/tf/kera s/applications/mobilenet/MobileNet
Publication Date
10/19/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
1
Total views since publication
PDF Downloads
5
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...