发明名称 IMAGE CLASSIFICATION APPARATUS, IMAGE CLASSIFICATION SYSTEM, IMAGE CLASSIFICATION METHOD, AND PROGRAM
摘要 PROBLEM TO BE SOLVED: To provide an image classification apparatus configured to generate an image classification rule by use of machine learning, an image classification system, an image classification method, and a program.SOLUTION: An image classification apparatus includes: feature quantity deriving means which derives feature quantity of each of document images with first classification labels added thereto; clustering processing execution means which executes clustering processing by use of the feature quantity, to calculate a joining distance for the document images to converge into one cluster; classification label application means which divides the cluster into a plurality of sub-clusters at a predetermined threshold of joining distance, to apply a second classification label to a document image included in each of the sub-clusters; and classification rule generation means which executes machine learning by use of the feature quantity and the second classification label, to generate a classification rule for classifying a document image having a feature quantity corresponding to the second classification label to a category specified by the first classification label.SELECTED DRAWING: Figure 6
申请公布号 JP2016071412(A) 申请公布日期 2016.05.09
申请号 JP20140196722 申请日期 2014.09.26
申请人 CANON INC 发明人 ABE NOBORU
分类号 G06F17/30;G06T7/00 主分类号 G06F17/30
代理机构 代理人
主权项
地址