Image classification systems based on CNN based IC

专利名称：Image classification systems based on CNN

based IC and light-weight classifier

发明人：Lin Yang,Patrick Z. Dong,Charles Jin

Young,Jason Dong,Wenhan Zhang,BaohuaSun

申请号：US15963990申请日：20180426公开号：US10402628B2公开日：20190903

专利附图：

摘要：Image classification system contains a CNN based IC configured for extracting

features out of input data by performing convolution operations using filter coefficientsof ordered convolutional layers and a classifier IC configured for classifying the inputdata using reduced set of the extracted features based on a light-weight classifier. Light-weight classifier is derived by: training filter coefficients of the ordered convolutionallayers using a dataset containing N labeled data, the trained filter coefficients are for theCNN based IC; outputting respective extracted features of the N labeled data afterperforming convolution operations of ordered convolutional layers using the trainedfilter coefficients, each labeled data contains X features; creating the reduced set of theextracted features by eliminating those of the X features that contain zeros in at least Mof the N labeled data; and adjusting M until the light-weight classifier achievessatisfactory results using the reduced set.

申请人：GYRFALCON TECHNOLOGY INC.

地址：Milpitas CA US

国籍：US

代理机构：RC Patent Services

更多信息请下载全文后查看