Skip to main content

Document details - Efficient Low Dimensional Face Recognition Method based on Salient Features built by GIST in Conjunction with DWT

Journal Volume 10, Issue 4, April 2021, Article 16393199 Vasantha Kumara M, Mohammed Rafi , " Efficient Low Dimensional Face Recognition Method based on Salient Features built by GIST in Conjunction with DWT" , International Journal of Application or Innovation in Engineering & Management (IJAIEM) , Volume 10, Issue 4, April 2021 , pp. 008-016 , ISSN 2319 - 4847.

Efficient Low Dimensional Face Recognition Method based on Salient Features built by GIST in Conjunction with DWT

    Vasantha Kumara M, Mohammed Rafi

Abstract

ABSTRACT The wearable technologies coupled with biometrics, are used to detect human beings effectively and also address their health issues. In this research paper, we propose a novel face recognition scheme termed Discrete Wavelet Transform (DWT) multiscale GIST (DWT-GIST) feature extraction with fewer number of effective features to escalate speed of the biometric system. The original face images of various datasets with different sizes are converted into uniform size and DWT is used to convert spatial domain into frequency bands. The low frequency band of DWT, which is one fourth dimension of the original face image is considered and further processed by using GIST to derive final low dimensional salient features.The K-Nearest Neighbors (K-NN) algorithm through seven distance equations are used to test and authenticate the face recognition system. The recognition rates of the proposed method are very promising and betterlikened to current methods. Keywords:Biometrics, DWT, Face Recognition, GIST, KNN

  • ISSN: 23194847
  • Source Type: Journal
  • Original language: English

Cited by 0 documents

Related documents

{"topic":{"name":"Order Picking; AS/RS; Warehouses","id":5729,"uri":"Topic/5729","prominencePercentile":98.30173,"prominencePercentileString":"98.302","overallScholarlyOutput":0},"dig":"7972b85ca5bc948c1a2f0423f8150b186ec6bb8cf32afac11c4a324b8d78fb11"}