RECOGNITION OF HUMAN FACES BASED ON POPULATION VERIFIED REPRODUCIBLE MEASUREMENTS BETWEEN FACIAL ANTHROPOLOGICAL LANDMARKS ON 2 DIMENSIONAL OR 3 DIMENSIONAL HUMAN PHOTOGRAPHS

摘要

A method of facial recognition has been developed by the application of a statistical method, standard deviations or standard errors versus sample number plots, to differentiate the degree of reproducibilities of various measurements between facial anthropological landmarks in individual ethnic groups. Reproducible measurements between facial anthropological landmarks in a particular ethnic group mean they are common features shared by individuals of that ethnic group. Non-reproducible measurements are unique features of each individual in that ethnic group which may be used for individual facial recognition purposes. Such methodology may be computerised for automatic facial recognition. Big data of each ethnic group are needed for facial recognition. In turn, the development of data bases of each ethnic group will result in big data of human faces.

主权项

1. A method of recognition of human faces based on population verified reproducibility of making linear and/or angular measurements between facial anthropological landmarks on 2 dimensional or 3 dimensional human photographs, comprising steps of:
Randomly collect male and female individuals from a population, taking 2-D or 3-D photographs of their faces, sorting these 2-D or 3-D photographs by their genders, identifying all these anthropological facial landmarks on 2-D or 3-D human photographs in male and female and making linear and/or angular measurements between anthropological facial landmarks. The said each linear measurement is measured between at least 2 anthropological landmarks. The said each angular measurement is measured between at least 3 anthropological landmarks. The said each repeated linear or angular measurement starting from the first (n=1) and second (n=2) individual is calculated for their means, standard deviations (SD) and standard errors (SE). The said every additional repeated measurement made on study subjects, i.e. n=1, 2, 3 followed by n=1, 2, 3, 4, etc., new sets of mean, SD and SE are calculated. The said cumulative standard deviation (SD) and/or cumulative standard errors (SE) are plotted against the number of individuals (n) measured for that particular linear or angular measurement.