Research on Gender-related Fingerprint Features, Extracting Fingerprint Features Using Autoencoder Networks for Gender Classification

Abstract

Background: Fingerprint is an important biological feature of human body, which contains abundant biometric information.At present, the academic exploration of fingerprint gender characteristics is generally at the level of understanding, and the standardization research is quite limited. Methods: A robust approach is presented in this article, Dense Dilated Convolution ResNet Autoencoder, to extract valid gender information from fingerprints. By replacing the normal convolution operations with the atrous convolution in the backbone, prior knowledge is provided to keep the edge details and the global reception field can be extended. Results: The results were explored from three aspects: 1) The efficiency of the DDC-ResNet. 6 typical automatic feature extraction methods with 9 mainstream classifiers for a total of 54 combinations are evaluated on our dataset and provide unbiased experimental details. Experimental results demonstrate that the combination of our approach outperforms other combinations in terms of average accuracy and separate-gender accuracy. It reaches 96.5% for average and 0.9752(males)/0.9548(females) for separate-gender accuracy. 2) The effect of fingers. It is found that the best performance of classifying gender with separate fingers is achieved by the right ring finger. 3) The effect of specific features. Based on the observations of the concentrations of fingerprints visualized by our approach, it can be inferred that loops and whorls (level 1), bifurcations(level 2), as well as line shapes (level 3) are connected with gender. Finally, we will open source the dataset that contains 6000 fingerprint images. Conclusions: The results are demonstrated that autoencoder networks are a powerful method for extracting gender-specific features to help hide the privacy information of the user’s gender contained in the fingerprint. Our experiments also identified three levels of features in fingerprints that are important for gender differentiation, including loops and whorls shape, bifurcations shape, and line shapes.

An Investigation of Benford’s Law Divergence and Machine Learning Techniques for Intra-Class Separability of Fingerprint Images

“…Baştürk et al (2018) recognised different types of fingerprint images using different machine learning techniques and observed that the deep neural network was more suited for the effective recognition of fingerprint images. Lastly, our proposed method when considering the CNN (99.72%) outperformed that of Qi et al (2022). Qi et al (2022) performed a gender-related classification based on fingerprint images using dense dilated convolution ResNet Autoencoder and achieved an average accuracy of 96.5%.…”

“…Lastly, our proposed method when considering the CNN (99.72%) outperformed that of Qi et al (2022). Qi et al (2022) performed a gender-related classification based on fingerprint images using dense dilated convolution ResNet Autoencoder and achieved an average accuracy of 96.5%.…”

SoftVein-WELM: A Weighted Extreme Learning Machine Model for Soft Biometrics on Palm Vein Images

“…From a practical point of view, gender classification and age estimation offer multiple advantages as soft biometrics are used to filter the database. In this regard, different authors have proposed different recognition approaches based on other biometric traits such as fingerprints [27,28], palmprints [29,30],…”

“…For gender classification, our method performs similarly to the best results on palmprint [30], which contains rich information also capturing some palm vein structures in an ROI area similar to palm veins. Compared to more traditional biometric methods, such as fingerprints, the iris, and the face, our method demonstrates simplicity based on W2-ELM contrasting hybrid [31,33,52], multi-task [35,53] or deep [27,33] models. On the other hand, age group classification is a less challenging task in vein-based biometrics, obtaining higher results than other traits.…”

review on identification of gender using fingerprints

“…The DDC-ResNet coupled with CNN to test the 10 sets the result indicates that for each specific finger, the right ring finger providesthe highest accuracy, reaching 92.455%. [1] (Shima Jalali 2021)Here, the author performed operation on real-time identification of fingerprints i.e.real fingerprints were captured from crime scenes, this captured fingerprint consists of noise,and thereforetheremulti-stage preprocessing was necessary so that preprocessing stages were consists of segmentation, normalization, median filter, binarization and thinning. After this stage, gender-related features were extracted by 5 efficient features for fingerprints, such as RTVTR, LBP, Entropy, DCT, Minutiae points then this all followed by classification stage as kNN, Support Vector Machine (SVM), random forest, Adaboost, Linear Discriminant Analysis (LDA) and a one hidden layer neural networkclassifiers separately.…”