首页>
外国专利>
PEDESTRIAN DETECTION METHOD USING HOG-PCA FEATURE AND RBFNNS PATTERN CLASSIFIER
PEDESTRIAN DETECTION METHOD USING HOG-PCA FEATURE AND RBFNNS PATTERN CLASSIFIER
展开▼
机译:HOG-PCA特征和RBFNNS模式分类器的行人检测方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
The present invention relates to a pedestrian detection method using a histogram of oriented gradient-principal component analysis (HOG-PCA) feature and an RBFNNs pattern classifier. The method comprises the following steps of: generating an HOG feature vector of an input image for study by using an HOG algorithm; reducing dimensions of the HOG feature vector by using a PCA algorithm; learning a feature of the input image for study; and detecting a pedestrian. Therefore, the method can rapidly and efficiently detect a pedestrian.;COPYRIGHT KIPO 2016
展开▼