Automatic Images classification Based on Multi-features Combined with MIL

摘要：

This paper presents a novel method for image classification. For classification, use a multiple instance learning (MIL) framework to describe the features as instances. By using a discriminative approach, we embeded each image into instance-based space and the coordinates of the image in that space are used for classification using the k-nearest neighbor algorithm. Expriments shows that the approac is usefulf for image cassification.

关键词： imag eclassiftcation MIL framework KNN

作者: Yang Liu Tiegang Bai

作者单位: College of Automation Engineering Shenyang Aerospace University Shenyang, China Police Dog Technique SchoolMinitry of Public Security Shenyang, China

会议类型: 国际会议

会议名称: 2011 4th International Congress on Image and Signal Processing(第四届图像与信号处理国际学术会议 CISP 2011)

会议地点: 上海

会议语种:英文

页码: 118-121

在线出版日期: 2011-10-15（万方平台首次上网日期，不代表论文的发表时间）