公开(公告)号：US20140348399A1

公开(公告)日：2014-11-27

申请号：US14252792

申请日：2014-04-15

Applicant: ASUSTeK COMPUTER INC.

Inventor： Hsiu-Jui KUO ,  Yu-Cheng LIU ,  Yi-Chi CHENG

IPC: G06T5/00 ,  G06K9/00

CPC classification number: G06K9/00288 ,  G06K9/00228

Abstract: An image processing method includes following steps; capturing an image including a human face; executing a face detection in a predetermined area of the image; adjusting an exposure value of the predetermined area to an expected value when the face detection does not recognize the human face; executing the face detection and analyzing an image information in a face area of the image when the exposure value of the predetermined area reaches the expected value; and selecting a model parameter and an image adjusting parameter correspondingly according to the image information, processing the face area of the image, and outputting a processed image to a human face recognition system.

Abstract translation: 一种图像处理方法包括以下步骤： 拍摄包括人脸的图像; 在图像的预定区域中执行面部检测; 当所述面部检测不识别人脸时，将所述预定区域的曝光值调整为预期值; 当预定区域的曝光值达到期望值时，执行面部检测和分析图像的面部区域中的图像信息; 以及根据图像信息相应地选择模型参数和图像调整参数，处理图像的面部区域，并将处理的图像输出到人脸识别系统。