Evaluation of face recognition techniques using PCA, wavelets and SVM

Gumus, Ergun; Kılıç, Niyazi; Sertbaş, Ahmet; Uçan, Osman N.

doi:10.1016/j.eswa.2010.02.079

Cited by 177 publications

(67 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Selain itu juga ada banyak penelitian yang menerapkan PCA untuk pengenalan wajah menggunakan eigenface ( [7], [14], [15]). Untuk Wavelet sebelumnya beberapa peneliti menerapkan wavelet pada ekstraksi fitur dan membandingkan dengan beberapa basis wavelet ( [5], [16], [17], [18]) dan ada beberapa penelian yang membandingkan ekstraksi fitur pada pengenalan wajah menggunkan PCA dan Wavelet ( [19], [20] …”

Section: Pendahuluanunclassified

Studi Komparasi Ekstraksi Fitur pada Pengenalan Wajah Menggunakan Principal Component Analysis (PCA) dan Wavelet Daubechies

Imah²

2015

JMASIF

View full text Add to dashboard Cite

AbstrakPaper ini membahas perbandingan ekstraksi fitur untuk pengenalan wajah menggunakan metode Principal Component Analysis (PCA) dan Wavelet Daubechies untuk pengenalan wajah . Basis waveletdaubechiesyang digunakan adalah wavelet db2, db4, dan db8. Setiap dekomposisi dilakukan hingga level ke-3 yang kemudian diambil fitur aproksimasiwavelet dan fitur statistik wavelet. Variasi nilai komponen utama dimulai dari nilai komponen ke-1 hingga nilai komponen ke-100 dari 4096 nilai eigen. Nilai komponen ke-1 memiliki presentasesebesar 62% sedangkan nilai komponen ke-100 memiliki presentase sebesar 99%dari total nilai eigen,. Pengujian sistem menggunakan 216 citra wajah yang diambil dari dataset The Japanese Female Facial Expression (JAFFE) yang terdiri dari 10 individu dengan masing-masing sekitar 20 wajah per-individu. Pemilihan data train dan data tes menggunakan cross validation dengan rata-rata akurasi 94.42%. Dari hasil percobaan menggunakan Random Forest Classifier diperoleh tingkat pengenalan tertinggi untuk ekstraksi menggunakan PCA sebesar 100% pada variasi data 95% ,sedangkan tingkat pengenalan tertinggi untuk ekstraksi menggunakan Wavelet Daubechies sebesar 98.611% pada wavelet db2 menggunakan fitur aproksimasi wavelet.Kata kunci : ekstraksi fitur, PCA, pengenalan wajah, Random Forest Classifier,Wavelet Daubechies.

show abstract

Section: Pendahuluanunclassified

Studi Komparasi Ekstraksi Fitur pada Pengenalan Wajah Menggunakan Principal Component Analysis (PCA) dan Wavelet Daubechies

Imah²

2015

JMASIF

View full text Add to dashboard Cite

show abstract

“…We remark that recognition accuracy rate of 0.9 is deemed as adequate for real-world face recognition applications [BO08,GKS10] and, following (5.1), our results can be derived for arbitrary recognition accuracy rates.…”

Section: Face Recognition On the Cloudmentioning

confidence: 99%

Bandit framework for systematic learning in wireless video-based face recognition

Atan

Tekin

Schaar

et al. 2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Professor Mihaela van der Schaar, Chair Video-based object or face recognition services on mobile devices have recently garnered significant attention, given that video cameras are now ubiquitous in all mobile communication devices. In one of the most typical scenarios for such services, each mobile device captures and transmits video frames over wireless to a remote computing cluster (a.k.a. "cloud" computing infrastructure) that performs the heavy-duty video feature extraction and recognition tasks for a large number of mobile devices. A major challenge of such scenarios stems from the highly-varying contention levels in the wireless transmission, as well as the variation in the task-scheduling congestion in the cloud. In order for each device to adapt the transmission, feature extraction and search parameters and maximize its object or face recognition rate under such contention and congestion variability, we propose a systematic learning framework based on multi-user multi-armed bandits. The performance loss under two instantiations of the proposed framework is characterized by the derivation of upper bounds for the achievable short-term and long-term loss in the expected recognition rate per face recognition attempt against the "oracle" solution that assumes a-priori knowledge of the system performance under every possible setting. Unlike well-known reinforcement learning ii techniques, such as Q-learning, that exhibit very slow convergence when operating in highly-dynamic environments, the proposed bandit-based systematic learning quickly approaches the optimal transmission and cloud resource allocation policies based on feedback on the experienced dynamics (contention and congestion levels).To validate our approach, time-constrained simulation results are presented via: (i) contention-based H.264/AVC video streaming over IEEE 802.11 WLANs and (ii) principal-component based face recognition algorithms running under varying congestion levels of a cloud-computing infrastructure. Against state-ofthe-art reinforcement learning methods, and for the same recognition accuracy rate set for each face recognition transaction between a mobile and the cloud, our bandit-based framework is shown to provide 17.8% ∼ 44.5% reduction of the number of video frames that must be processed by the cloud for recognition and 11.5% ∼ 36.5% reduction in the video traffic over the WLAN.iii The thesis of Onur Atan is approved.

show abstract

“…SVM was proposed by Vapnik in 1998 [10]. The hyperplane is found in SVM in which the separation of the largest possible fraction of points of the same class on the same side is obtained and maximization of the distance from either class to the hyperplane is carried out [11].…”

Section: Support Vector Machinementioning

confidence: 99%

“…Separation of two data sets is carried out by searching for an optimal separating hyperplane (OSH) between them in SVM. ''Support vectors" are the bounds between data sets and OSH [10]. The data, that are not linearly separable, are transformed into new space using kernel and OSH is found.…”

Section: Support Vector Machinementioning

confidence: 99%

A New Paradigm of Face Recognition Using Multi-class Classifier

Shah¹

2017

IJRASET

View full text Add to dashboard Cite

Eminent applications for security and law enforcement have enhanced significant importance of automatic face recognition. Face recognition is one of the most convenient ways for identification and recognition of a person in considerable large database. The face recognition complications are due to change in illumination of light sources, variations in pose as well as facial expressions. This paper will present a novel approach of hybrid face recognition technique. The face recognition system is developed with PCA and SVM as multi class classifier. The experiments of the proposed method have been conducted on ORL data sets. The experiment result is presented and discussed, which shows the effectiveness of the strategy described.

show abstract

Evaluation of face recognition techniques using PCA, wavelets and SVM

Cited by 177 publications

References 4 publications

Studi Komparasi Ekstraksi Fitur pada Pengenalan Wajah Menggunakan Principal Component Analysis (PCA) dan Wavelet Daubechies

Studi Komparasi Ekstraksi Fitur pada Pengenalan Wajah Menggunakan Principal Component Analysis (PCA) dan Wavelet Daubechies

Bandit framework for systematic learning in wireless video-based face recognition

A New Paradigm of Face Recognition Using Multi-class Classifier

Contact Info

Product

Resources

About