Several LBPs were designed for texture description, and induced

Several local feature descriptors were introduced for facial
image analysis. Among those LBP, LGP, LTP, CENTRIST and
NABP have been proposed for the classification of images. The
techniques of those descriptors are traced in this section.
Among all the feature descriptors, Ojala et al. first introduce
original LBP operator which thresholds n×n neighborhood of
every pixel of an image with the center pixel value and considers
the result as a binary number 31. LBPs were designed
for texture description, and induced for face representation
in several applications including face detection 19, 11,
18, face recognition 46, 1, 54, flower classification49,
object classification , leaf classification , scene classification,
expression recognition 41, 10, 44, gender classification
36, 45 and texture classification 43.
LBP is additionally utilized by Shan et al. 41 for expression
recognition. They divided the face image into many
subregions of different size and extracted options from solely
few sub-regions which are classified using boosted SVM 3
afterwards. Beyond that, real time gender recognition using
boosted LBP features was sought by C. Shan 40 . Inspired
by the tremendous performance of LBP operator, Tan and
Triggs 46 amplify the binary pattern into ternary pattern
which encodes facial images using a fixed threshold (±5).
Although LBP has gained popularity for its simplicity, it
fails to differentiate a small difference and a large difference
in acuities which deteriorates its preferential capacity. Another
weakness of LBP is that it can be affected by the noise due
to local intensity fluctuation especially in uniform and near
uniform regions.
Jun et al. 19 made an exploration to use adaptive threshold
in LBP and proposed LGP for face and human detection
where n × n neighborhood of a pixel is considered, and the
neighbor having gradient greater than or equal to the average
of gradients of eight neighboring pixels, is set to a binary value
of “1”, otherwise is assigned a binary value of “0”.
We et al. 49 introduced CENTRIST which is a visual
feature descriptor for scene and object classification which
performs a census transform (CT) of an image and replaces
the image with its CT values 49. CT is a non-parametric local
transformation designed for establishing relationships between
local patches 52, which is calculated similarly as LBP.
Rahman et al. 37 proposed Noise Adaptive Binary Pattern
(NABP) for facial image analysis such as face recognition,
expression recognition and gender classification. NABP encodes
the face microstructures using an adaptive threshold and
generates more discriminative patterns than other existing local
feature descriptors.
The local feature descriptors so far we depicted were proposed
for different applications on image analysis. However
more research should be conducted to increase the result of
the accuracy for these applications.