An Improved LBP Blockwise Method for Face Recognition

An Improved LBP Blockwise Method for Face Recognition

Nikhil Kumar, Sunny Behal
Copyright: © 2018 |Pages: 11
DOI: 10.4018/IJNCR.2018100103
(Individual Articles)
No Current Special Offers


Face recognition is considered as one of toughest and most crucial leading domains of digital image processing. The human brain also uses a similar kind of technique for face recognition. When scrutinizing a face, the human brain signifies the result. Aside from AN automatic processing system, this technique is very sophisticated, owing to the image variations on account of the picture varieties in as far as area, size, articulation, and stance. In this article, the authors have used the options of native binary pattern and uniform native binary pattern for face recognition. They compute a number of classifiers on publicly available benchmarked ORL image databases to validate the proposed approach. The results clearly show that the proposed LBP-piece shrewd strategy has outperformed the traditional LBP system.
Article Preview

1. Introduction

Face recognition could be a subject beneath analysis for pretty much a recent decade. Automatic recognition could be an intimidating task. There are numerous applications of this field such as banks, high security official offices, business store, malls, personal firms, etc.

In any case, confront acknowledgment information can be inclined to mistake, which can involve individuals for wrongdoings they haven't submitted. Facial acknowledgment programming is especially awful at perceiving African Americans and other ethnic minorities, ladies, and youngsters, frequently misidentifying or neglecting to distinguish them, dissimilarly affecting certain gatherings. Also, confront acknowledgment has been utilized to target individuals taking part in ensured discourse. Sooner instead of later, confront acknowledgment innovation can most likely prove to be additional present. It might be utilized to track people's developments out on the planet like mechanized tag peruses track vehicles by plate numbers. Constant face acknowledgment is now being utilized as a part of different nations and even at brandishing occasions in America.

Face acknowledgment frameworks change in their capacity to distinguish individuals under testing conditions, for example, poor lighting, low quality picture determination, and problematic point of view, (for example, in a photo brought from above looking down on an obscure individual) (Li & Lu, 1999; Hsu & Abdel-Mottaleb, 2002; Rahman & Afrin, 2013; Satapathy et al., 2016; Bhateja, Tavares, Rani, Prasad, & Raju, 2018).

To keep the mistakes in check, two important concepts are there to realize:

  • An error statement “false negative” will be appear as the system failed to recognize the face of a person in image database thus return no result to the query;

  • An error statement “false positive” will be appear in case of successful identification of a person but the match is false (Chidambaranathan, 2015).

For instance, in the event that you are utilizing face acknowledgment to open your smartphone, it is better if the framework neglects to distinguish you a couple of times (false negative) than it is for the framework to misidentify other individuals as you and gives those individuals a chance to open your phone (false positive). On the off chance that the aftereffect of a wrong recognizable proof is that a pure individual goes to imprison (like a wrong ID in a mug shot database), at that point the framework ought to be intended to have a couple of false positives as could be allowed.

There are two terms used in case of face recognition i.e. face recognition and face detection. There is crystal clear difference between terms detection and recognition. Face detection works to find a face in an image whereas face recognition comes into action to disclose the identity of the face. Face recognition is next step of face detection (Singh, Sharma, & Rao, 2011; Li & Lu, 1999; Roy & Podder, 2013; Dutta & Baru, 2013).

Figure 1.

A typical face recognition system


Complete Article List

Search this Journal:
Volume 12: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing