期刊名称：International Journal of Artificial Intelligence & Applications (IJAIA)
印刷版ISSN：0976-2191
电子版ISSN：0975-900X
出版年度：2019
卷号：10
期号：4
页码：1-9
DOI：10.5121/ijaia.2019.10404
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Image classification is a popular machine learning based applications of deep learning. Deep learningtechniques are very popular because they can be effectively used in performing operations on image datain large-scale. In this paper CNN model was designed to better classify images. We make use of featureextraction part of inception v3 model for feature vector calculation and retrained the classification layerwith these feature vector. By using the transfer learning mechanism the classification layer of the CNNmodel was trained with 20 classes of Caltech101 image dataset and 17 classes of Oxford 17 flower imagedataset. After training, network was evaluated with testing dataset images from Oxford 17 flower datasetand Caltech101 image dataset. The mean testing precision of the neural network architecture withCaltech101 dataset was 98 % and with Oxford 17 Flower image dataset was 92.27 %..
关键词：Image Classification; CNN; Deep Learning; Transfer Learning
