Deep Learning Image Classification Based on Neural Network Optimized SVM
Download as PDF
DOI: 10.23977/icamcs.2018.062
Author(s)
Na Chen, Aiping Xiao, Gang Zheng
Corresponding Author
Na Chen
ABSTRACT
The Image classification method, which based on deep learning, can learn hierarchical feature description in supervised or unsupervised way and thus replace the manual design or selection of image features. Convolution Neural Network (CNN) in deep learning model has made remarkable achievements in the field of image in recent years. Directly using image pixel information as input, CNN can retain all information of the image to the greatest extent, then the recognition results can be given out through output models after convolution operations of feature extraction and high-level abstraction. This direct end-to-end learning method based on "input-output" has made great achievements and has been widely used.
KEYWORDS
Convolution neural network (CNN), image classification, SVM, deep learning