Face recognition apparatus, face recognition method, gabor filter application apparatus, and computer program

A method for a face recognition apparatus of recognizing an input face image based on one or more registered face images, comprising:

• inputting the face image to be recognized;
• calculating, at a processor of the face recognition apparatus by applying a Gabor filter, convolution of a pixel value at each feature value extraction position of the input image and coefficients of a Gabor filter, the coefficients of the Gabor filter being defined by a Gabor function representing a spatial characteristic using a Gaussian function representing a window and a sine function or a cosine function representing a frequency response, for each of a predetermined set of frequency values and for each of a predetermined set of angles of a response function of the Gabor filter thereby determining a Gabor jet having as many scalar values as there are Gabor filters;
• calculating a similarity between a Gabor jet at each feature value extraction position of the input face image and a Gabor jet of a registered face image;
• determining a similarity vector whose elements are given by values indicating similarity at respective feature value extraction positions;
• and
• determining when the input face image is identical to the registered face image, based on the similarity vector,
• wherein the step of applying the Gabor filter further includes
• reading values from a
• filter window data ROM that stores values of the filter window determined by calculating the Gaussian function,
• sine data ROMs that store values of the response function determined by calculating the sine function for the respective angles, and
• cosine data ROMs that store values of the response function determined by calculating the cosine function for the respective angles, and
• calculating
• coefficients of the Gabor filter by multiplying a value read from the filter window data ROM and the values read from the sine data ROMs and the cosine data ROMs, and
• a convolution of the input face image and a convolution of the calculated coefficients of the Gabor filter.

A Gabor filter application apparatus comprising:

• a Gabor filter application unit, implemented via a processor, that calculates convolution of spatial data and coefficients of a Gabor filter, the coefficients of the Gabor filter being defined by a Gabor function representing a spatial characteristic using a Gaussian function representing a window and a sine function or a cosine function representing a frequency response, for each of a predetermined set of frequency values and for each of a predetermined set of angles of a response function of the Gabor filter thereby determining a Gabor jet having as many scalar values as the number of Gabor filter types, the Gabor filter application unit including
• a filter window data ROM that stores values of the filter window determined by calculating the Gaussian function,
• sine data ROMs that store values of the response function determined by calculating the sine function for the respective angles,
• cosine data ROMs that store values of the response function determined by calculating the cosine function for the respective angles, and
• a Gabor filter coefficient calculation unit that calculates coefficients of the Gabor filter by multiplying a value read from the filter window data ROM and values read from the sine data ROMs and the cosine data ROMs, and
• a convolution calculating unit that calculates the convolution using the Gabor filter coefficients calculated by the Gabor filter coefficient calculation unit.

A non-transitory computer-readable medium storing computer readable instructions thereon for recognizing an input face image based on one or more registered face images, the computer readable instructions which when executed by a face recognition apparatus cause the face recognition apparatus to perform a method comprising:

• inputting the face image to be recognized;
• calculating, by applying a Gabor filter, convolution of a pixel value at each feature value extraction position of the input image and coefficients of a Gabor filter, the coefficients of the Gabor filter being defined by a Gabor function representing a spatial characteristic using a Gaussian function representing a window and a sine function or a cosine function representing a frequency response, for each of a predetermined set of frequency values and for each of a predetermined set of angles of a response function of the Gabor filter thereby determining a Gabor jet having as many scalar values as there are Gabor filters;
• calculating a similarity between a Gabor jet at each feature value extraction position of the input face image and a Gabor jet of a registered face image;
• determining a similarity vector whose elements are given by values indicating similarity at respective feature value extraction positions;
• determining when the input face image is identical to the registered face image, based on the similarity vector,
• wherein the step of applying Gabor filter further includes
• reading values from a
• filter window data ROM that stores values of the filter window determined by calculating the Gaussian function,
• sine data ROMs that stores values of the response function determined by calculating the sine function for the respective angles, and
• cosine data ROMs that stores values of the response function determined by calculating the cosine function for the respective angles, and
• calculating
• coefficients of the Gabor filter by multiplying a value read from the filter window data ROM and the values read from the sine data ROMs and the cosine data ROMs, and
• a the convolution of the input face image and a convolution of the calculated coefficients of the Gabor filter.

A face recognition apparatus for recognizing an input face image based on one or more registered face images, comprising:

• face image input means for inputting the face image to be recognized;
• Gabor filter application means for calculating convolution of a pixel value at each feature value extraction position of the input image and coefficients of a Gabor filter, the coefficients of the Gabor filter being defined by a Gabor function representing a spatial characteristic using a Gaussian function representing a window and a sine function or a cosine function representing a frequency response, for each of a predetermined set of frequency values and for each of a predetermined set of angles of a response function of the Gabor filter thereby determining a Gabor jet having as many scalar values as there are Gabor filters;
• similarity calculation means
• for calculating the similarity between a Gabor jet at each feature value extraction position of the input face image and a Gabor jet of a registered face image, and
• for determining a similarity vector whose elements are given by values indicating similarity at respective feature value extraction positions;
• and
• identification determination means for determining when the input face image is identical to the registered face image, on the basis of the similarity vector,
• the Gabor filter application means including
• filter storing means for storing values of the filter window determined by calculating the Gaussian function,
• sine value storing means for storing values of the response function determined by calculating the sine function for the respective angles,
• cosine value storing means for storing values of the response function determined by calculating the cosine function for the respective angles, and
• Gabor filter coefficient calculation means for calculating
• coefficients of the Gabor filter by multiplying a value read from the filter storing means and the values read from the sine value storing means and the cosine value storing means, and
• a convolution of the input face image and a convolution of the Gabor filter coefficients calculated by the Gabor filter coefficient calculation unit.