Method of and system for image processing and computer program

Abstract

To provide an image processing method that detects the width of a face imaged in a face photograph image. A skin color area extraction part 70 extracts a skin color area from a face image S1 segmented from a face photograph image S0. A face width acquisition part 90 scans the image of the skin color area in a fast scan direction and a slow scan direction that are a lateral direction (a direction of extension of a line connecting both eyes) and longitudinal direction (a direction perpendicular to the lateral direction) of the face respectively to detect the width of the area in each slow scan position, and according to the mode of width changes along the longitudinal direction, detects the slow scan position where the width increases discontinuously as a first position and the slow scan position that is below the first position and just above the slow scan position where the width reduces discontinuously as a second position. Out of the widths in the slow scan positions in the range from the first position to the second position, the largest width is decided as the width W of the face.

Description

technical field [0001] The present invention relates to an image processing, in particular to an image processing method and device for detecting the width of a captured face in a facial photograph image and a program for the same. Background technique [0002] In the application for approval of passports and driver's licenses, or the production of resumes, it is often required to submit a photo of the intended output standard with the face of the person (hereinafter referred to as an ID photo). The output specifications of ID photos almost always stipulate the length of the entire completed image and the length of the face (or part of the face) in the vertical direction, while the length of the entire completed image in the left-right direction ( Width) is regulated, and there is no regulation on the width of the face. [0003] In order to obtain a photo of an ID like this, various methods have been proposed. For example, as described in Patent Document 1, it is proposed ...

AI Technical Summary

Problems solved by technology

[0007] In this way, in order to create a photograph that specifies the width of the face in the completed image, it is necessary to grasp the width of the face in the original facial photograph image, but there is no method to detect the width of the face captured in the facial photograph image

Images