Face detection and tracking using MATLAB

Master's Thesis, 2014

10 Pages, Grade: 3


Face Detection and Tracking Using MATLAB

Onkar Nath Sabran

Department of Electronics and Communication Engineering

Mtech in VLSI

Abstract—Face Identification and following has been a vital and dynamic examination field on the grounds that it offers numerous requisitions, particularly in feature observation, biometrics, or feature coding. The objective of this undertaking was to actualize a constant framework on a FPGA board to catch and track a human’s face. The face location calculation included shade based skin division and picture separating. The face area was dictated by figuring the centroid of the discovered locale. A product variant of the calculation was autonomously executed and tried on still pictures in MATLAB. Despite the fact that the move from MATLAB to verilog was not as smooth obviously, trial results demonstrated the exactness and viability of the constant framework, much under shifting states of lights, facial postures and skin colors, All estimation of the fittings usage was carried out continuously with negligible computational exertion consequently suitable for force constrained provisions.

I. Introduction

Face location and following is the methodology of figuring out if or not a face is available in picture. Dissimilar to face distinguishment which recognizes distinctive human confronts, face location just demonstrates whether a face is available in a picture. Moreover, face following decides the precise area of the face. Face identification and following has been a dynamic examination range for quite a while in light of the fact that it is the starting imperative venture in numerous distinctive requisitions, for example, feature observation, face distinguishment, picture upgrade, feature coding, and vitality protection. The materialness of fae identification in vitality preservation is not as evident as in different provision. Not with standing, it is fascinating to figure out how a face identification and following framework permits force and vitality to be spared. Assume one is viewing a TV and dealing with different errands all the while. The face location framework is for checking whether the individual is gazing toward straight forwardly toward the TV.



Abbildung in dieser Leseprobe nicht enthalten

General configuration stages are shown in Figure 1.

The skin location calculation here was determined from the technique describe in [1]. Color division has been turned out to be a compelling strategy to catch face is as because of its low computational prerequisites and simplicity of execution. Contrasted with the emphasized based strategy, the shade based calculation obliged almost no preparation. Initially, the first picture was changed over to an alternate color space, specifically altered YUV. At that point the skin pixels were portioned focused around the proper u reach. Morphological sifting was connected to diminish false positives. At that point each one joined locale of caught pixels in the picture was marked. The territory of each one marked locale was registered and a region based separating was connected. Just districts with extensive region were considered face areas. The centroid of each one face area was like wise processed to demonstrate to its area.


Excerpt out of 10 pages


Face detection and tracking using MATLAB
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
485 KB
face, matlab
Quote paper
Onkar Sabran (Author), 2014, Face detection and tracking using MATLAB, Munich, GRIN Verlag, https://www.grin.com/document/274960


  • No comments yet.
Read the ebook
Title: Face detection and tracking using MATLAB

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free