Palmprint identification is part of biometric recognition, which attracted many researchers, especially when fusion with face identification that will be applied in the airport to hasten knowing individual identity. To accelerate the process of verification feature palms, dimension reduction method is the dominant technique to extract the feature information of palms. The mechanism will boost if the ROI images are processed prior to get normalize image enhancement. In this paper with three sample input database, a kernel PCA method used as a dimension reduction compared with three others and a skeleton filter used as a image enhancement method compared with six others. The final results show that the proposed method successfully achieve the target in terms of the processing time of 0.7415 second, the EER performance rate of 0.19 % and the success of verification process about 99,82 %.
In recent years, biometric recognition has been rapidly developed and still continues to grow. Researchers are combining several algorithms to obtain a more robust feature. In this study, Gabor kernel methods, principle component analysis (PCA), detection error trade-off (DET), expected performance curves (EPC), and cumulative match characteristic (CMC) is combined and used to obtain the features of palm print. This experiment shows that the combination of Gabor and PCA methods, using 240 items of data, gives an optimum result in palms identification and authentication. Intisari-Pengenalan biometrik terus berkembang sehingga banyak peneliti menggabungkan beberapa algoritma dalam mendapatkan fitur yang lebih kokoh. Penelitian ini menggabungkan metode kernel Gabor dengan PCA, DET, EPC, dan CMC untuk memperoleh fitur dalam pengenalan telapak tangan. Hasil penelitian yang memberikan nilai optimal proses identifikasi dan authentikasi telapak tangan dari data sebanyak 240 item adalah pada metode penggabungan Gabor dan PCA.
Nowadays, the palmprint recognition method is an attractive tool for people recognition besides face recognition and fingerprint. To get a good palmprint recognition system requires a series of algorithms that are united and worked together to create a preferred method. The biometric system in palmprint recognition is said to be preferred when the choice of the algorithm applied has the advantage of each phase. Generally, The palmprint recognition system starts from the selection of image filters, image positioning, dimension reduction, and finally, end with the distance method. Image normalization position is a problem for researchers because the palmprint is always moving and shows a curved shape making it difficult for the system in the identification and verification process. For this reason, the research focuses on the selection of orientation scale operator value from the Gabor technique to get the best parameter value for palmprint biometrics. The desired system is when the overall performance of the algorithm used will produce an output with a low EER value and a high level of verification. From the research that has been carried out, the range of 8 × 5 becomes a good alternative for the choice of Gabor value pairs.
Palmprint Recognition Technology now requires breakthrough identification of diverse people. Palmprint recognition is the right choice of the system, namely the acceptance of biometrics that can be done quickly and cheaply because it has a significant enough media dimension that is difficult to manipulate. The four classifications of the biometric algorithm, the use of matching methods is still less attractive to researchers. In general, the part of the research concern is in the preprocessing and dimension reduction sections. Although researchers more widely use the Euclidean matching method, the selection of cosine methods is worth considering. The cosine method for the palmprint recognition matching process will have a significant effect on increasing the verification value when inserting the use of the contra-variance formulae in the equation. The selected amount of contra-variance is the data of the training. From the results of research that have been carried out, the rate of EER and verification are quite promising. The value of research results can compensate for other researchers in the same field.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.