简体   繁体   English

Haarcascade仅在348x288图像上运行?

[英]Haarcascade operates on 348x288 images only?

I am using opencv and c++. 我正在使用opencv和c ++。 Which face detector algorithm to use if I have 348x288 face images. 如果我有348x288的人脸图像,应使用哪种人脸检测器算法。 In the paper for Haarcascade http://www.vision.caltech.edu/html-files/EE148-2005-Spring/pprs/viola04ijcv.pdf , it is said that haarcascade operates on 348x288 pixel images. 在Haarcascade的论文http://www.vision.caltech.edu/html-files/EE148-2005-Spring/pprs/viola04ijcv.pdf中 ,据说haarcascade在348x288像素的图像上运行。 Does that mean I cannot use haarcascade to detect the faces in my images? 这是否意味着我无法使用级联来检测图像中的面部?

It can be used for your images as long as you setup the correct parameters for CascadeClassifier::detectMultiScale() , especially the following three: 只要为CascadeClassifier::detectMultiScale()设置正确的参数CascadeClassifier::detectMultiScale()尤其是以下三个参数CascadeClassifier::detectMultiScale() ,它就可以用于您的图像:

  • scaleFactor – Parameter specifying how much the image size is reduced at each image scale. scaleFactor –该参数指定每个图像比例缩小多少图像。

  • minSize – Minimum possible object size. minSize –最小可能的对象大小。 Objects smaller than that are ignored. 小于此值的对象将被忽略。

  • maxSize – Maximum possible object size. maxSize –最大可能的对象大小。 Objects bigger than that are ignored. 大于此值的对象将被忽略。

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM