BEAT - tpereira/tan

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.

Unnamed group

Endpoint Name	Data Format	Nature
image_gray	system/array_2d_uint8/1	Input
tantriggs_image	system/array_2d_floats/1	Output

Parameters 6

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

Name	Type	Default
sigma0	float64	1.0
sigma1	float64	2.0
gamma	float64	0.2
kernel_size	int64	5
threshold	float64	10.0
alpha	float64	0.1

import bob
import numpy

class Algorithm:

def __init__(self):
    self.sigma0      = 1
    self.sigma1      = 2
    self.gamma       = 0.2
    self.kernel_size = 5
    self.threshold   = 10.0
    self.alpha       = 0.1

def setup(self, parameters):
    self.sigma0      = parameters.get('sigma0', self.sigma0)
    self.sigma1      = parameters.get('sigma1', self.sigma1)
    self.gamma       = parameters.get('gamma', self.gamma)
    self.kernel_size = parameters.get('kernel_size', self.kernel_size)
    self.threshold   = parameters.get('threshold', self.threshold)
    self.alpha       = parameters.get('alpha', self.alpha)

self.m_tantriggs = bob.ip.TanTriggs(gamma=float(self.gamma), sigma0=float(self.sigma0), sigma1=float(self.sigma1), radius=int(self.kernel_size), threshold=float(self.threshold), alpha=float(self.alpha))

return True

def process(self, inputs, outputs):

image_gray      = inputs["image_gray"].data.value
    image_tantriggs = numpy.ndarray(image_gray.shape, numpy.float64)

# perform Tan&Triggs normalization    
    self.m_tantriggs(image_gray, image_tantriggs)

outputs["tantriggs_image"].write({
            'value': numpy.vstack(image_tantriggs)
        })

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 is a multi-stage image preprocessor relying on the Bob library. It implements the multi-stage preprocessing method described in [Tan07], the steps being a gamma correction, a difference of Gaussian filtering and a contrast equalization

Both inputs and outputs are expected to be grayscale images as two-dimensional arrays of floats (64 bits).

[Tan07]

Tan, B. Triggs: Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions. Analysis and Modeling of Faces and Gestures (2007) 168-182

Experiments

Attestation:

Privacy:

Status:

			Updated	Name	Databases/Protocols	Analyzers
				martabarrero/smarcel/full_isv/1/Prueba_ISV_2	banca/1@Mc	tutorial/eerhter_postperf/1

Scientific Python 2.7 (0.0.3) 55

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 tan_triggs 4

Endpoint Groups 1

Unnamed group

Parameters 6

Experiments

algorithms

tpereira

tan_triggs

4