This algorithm is a legacy one. The API has changed since its implementation. New versions and forks will need to be updated.

This algorithm is splittable

Endpoint Groups 1

Algorithms have at least one input and one output. All algorithm endpoints are organized in groups. Groups are used by the platform to indicate which inputs and outputs are synchronized together. The first group is automatically synchronized with the channel defined by the block in which the algorithm is deployed.

Group: main

Endpoint Name	Data Format	Nature
image	system/array_3d_uint8/1	Input
eye_centers	system/eye_positions/1	Input
lbp_histogram	system/array_1d_floats/1	Output

Parameters 9

Parameters allow users to change the configuration of an algorithm when scheduling an experiment

Name	Description	Type	Default	Range/Choices
left-eye-x	Final position of the left eye (x-coord)	uint32	44
left-eye-y	Final position of the left eye (y-coord)	uint32	12
color	Final color channel	string	gray	gray, red, green, blue, ichrominance
lbp-radius	Radius considered in the LBP bin computation	uint32	2
crop-height	Final image height	uint32	25
lbp-neighbours	Number of neighbours considered for the LBP bin computation	uint32	8
crop-width	Final image width	uint32	58
right-eye-x	Final position of the right eye (x-coord)	uint32	11
right-eye-y	Final position of the right eye (y-coord)	uint32	12

import bob
import numpy

class Algorithm:

def setup(self, parameters):
        self.crop_height = parameters.get('crop-height')
        self.crop_width  = parameters.get('crop-width')
        self.right_eye_y = parameters.get('right-eye-y')
        self.right_eye_x = parameters.get('right-eye-x')
        self.left_eye_y  = parameters.get('left-eye-y')
        self.left_eye_x  = parameters.get('left-eye-x')
        self.color       = parameters.get('color')

self.lbp_radius     = parameters.get('lbp-radius')
        self.lbp_neighbours = parameters.get('lbp-neighbours')

self.cropper = bob.ip.FaceEyesNorm(self.crop_height, self.crop_width,
                self.right_eye_y, self.right_eye_x, self.left_eye_y, self.left_eye_x)
                
        self.lbp_extractor = bob.ip.LBP(neighbors=int(self.lbp_neighbours), uniform=True, radius=float(self.lbp_radius), circular=True)

return True

def process(self, inputs, outputs):

image = inputs["image"].data.value

if (self.color == "gray"):
          image = bob.ip.rgb_to_gray(image)
        elif (self.color == "red"):
          image = image[0,:,:]
        elif (self.color == "green"):
          image = image[1,:,:]
        elif (self.color == "blue"):
          image = image[2,:,:]
        elif (self.color == "ichrominance"):
          image = 0.596 * image[0,:,:] - 0.275 * image[1,:,:] - 0.321 * image[2,:,:]
        else:
          raise ValueError("The filter {0} does not exists!!!".format(self.color))

eye_positions = inputs["eye_centers"].data
        cropped_image = self.cropper(image, float(eye_positions.right.y), float(eye_positions.right.x), float(eye_positions.left.y), float(eye_positions.left.x)).astype("uint8")

lbpimage = numpy.ndarray(self.lbp_extractor.get_lbp_shape(cropped_image), 'uint16') # allocating the image with lbp codes
        
        self.lbp_extractor(cropped_image, lbpimage) # calculating the lbp image
        features = bob.ip.histogram(lbpimage, 0, self.lbp_extractor.max_label-1, self.lbp_extractor.max_label).astype('float')
        features /= sum(features) #normalizing

outputs["lbp_histogram"].write({
            'value': features
        })

return True

The code for this algorithm in Python
The ruler at 80 columns indicate suggested POSIX line breaks (for readability).
The editor will automatically enlarge to accomodate the entirety of your input
Use keyboard shortcuts for search/replace and faster editing. For example, use Ctrl-F (PC) or Cmd-F (Mac) to search through this box

This algorithm performs the conversion of RGB images to: grayscale, red channel, green channel, blue channel or ichrominance of an image followed by a periocular cropping of the face, given the eye center coordinates.

The ichormiance conversion is the one implemented in [Lui2012] and is defined as follows:

I = 0.596*R − 0.275*G − 0.321*B, where R, G and B are respectively the red, green and blue channels.

This implementation relies on the `Bob &amp;amp;amp;amp;lt;http://www.idiap.ch/software/bob&amp;amp;amp;amp;gt;`_ library.

The inputs are:

image: an RGB image as a three-dimensional array of uint8, the first dimension being the number of color planes (3), and the second and third dimensions corresponding to the height and width of the original image, respectively.
eye_centers: the coordinates of the eye centers in the original image space

The output cropped_image is a grayscale cropped image as a two-dimensional array of floats (64 bits).

[Lui2012]

Lui, Yui Man, et al. &amp;amp;amp;quot;Preliminary studies on the good, the bad, and the ugly face recognition challenge problem.&amp;amp;amp;quot; Computer Vision and Pattern Recognition Workshops (CVPRW), 2012 IEEE Computer Society Conference on. IEEE, 2012.

Docutils System Messages

System Message: ERROR/3 (<string>, line 10); backlink

Unknown target name: "bob &amp;amp;amp;amp;lt;http://www.idiap.ch/software/bob&amp;amp;amp;amp;gt;".

Experiments

Attestation:

Privacy:

Status:

			Updated	Name	Databases/Protocols	Analyzers
				tpereira/tutorial/full_lbphs/2/btas2015_LBPBaseline_face_cpqd-smartphone-male_det	cpqd/1@smartphone_male	tutorial/eerhter_postperf_iso/1
				tpereira/tutorial/full_lbphs/2/btas2015_LBPBaseline_face_cpqd-smartphone-female_det	cpqd/1@smartphone_female	tutorial/eerhter_postperf_iso/1

This table shows the number of times this algorithm has been successfully run using the given environment. Note this does not provide sufficient information to evaluate if the algorithm will run when submitted to different conditions.

algorithms tpereira periocular_crop_lbphistogram 3

Endpoint Groups 1

Group: main

Parameters 9

Docutils System Messages

Experiments

algorithms

tpereira

periocular_crop_lbphistogram

3