Integrated Transfer Learning Method for Image Recognition Based on Neural Network

Abstract

As recognition technology advances in various fields, the problem of requiring a large number of correct image recognition is also accompanied. So as to promote the correctness of image recognition, we adopted two Convolutional Neural Network (CNN) integrated migration learning network models VGG-16 CNN and LeNet-5 CNN. Experimental results prove that the performance of image recognition in various models are improved. Also, the migration learning and convolutional neural network are used to construct a K-selection migration learning algorithm and a migration framework model whose parameters are migrated to the image data set. The use of a label data set with a large sample size, the selection of a pre-trained convolutional neural network model with a deep layer of complex network structure and the selection of an appropriate network fine-tuning depth can optimize the migration adaptation effect of the migration network. The image recognition precision of the algorithm is improved by about 3%, which further improves the migration adaptation effect.

KEYWORDS:

• Convolutional neural network
• K-selection transfer learning algorithm
• Image recognition
• LeNet-5 CNN
• Transfer learning
• VGG-16 CNN

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

JingYuan He

JingYuan He was born in Yan’an, Shaanxi, P R China, in 1987. She received the BS degree in School of Mathematics and Computer Science from Yan’an University in 2010, and the MS degree in School of Computer Science and Engineering from Xi'an Technological University in 2013, People’s Republic of China. She is currently pursuing the PhD degree at Rocket Force University of Engineering, and she is also a lecturer at Yan’an University. Her research interest include machine learning, image recognition and classification, and transfer learning.

BaiLong Yang

BaiLong Yang received his BS and MS degrees in computer applications technology from Rocket Force University of Engineering, Xi’an, People’s Republic of China, in 1990 and 1993, respectively. He received the PhD degree in aeronautics and astronautics manufacturing engineering from Rocket Force University of Engineering in 2001. He is currently a professor of Rocket Force University of Engineering. His research interests include image processing, network security, complex network and computer simulation. Email: [email protected]

Yang Su

Yang Su was born in Wuzhong, Ningxia, People’s Republic of China, in 1986. He received the BS and MS degrees in engineering of micro electronics and solid state electronics from Information Engineering University of PLA, in 2009 and 2012. He is currently pursuing the PhD degree at Rocket Force University of Engineering, and he is also a lecturer at Engineering University of People’s Armed Police. His research interests include image processing, fully homomorphic encryption hardware accelerator design and integrated circuit design. Email: [email protected]