Affiliations: [a] Department of Computer Science and Technology, Jilin University, Changchun, China
| [b] College of Information Technology and Media, Beihua University, Jilin, China
| [c] State Key Laboratory of Applied Optics, Changchun, China
| [d]
Changchun University of Science and Technology, Changchun, China
| [e] Department of Symbolic Computing and Knowledge Engineering, Key Laboratory of the Ministry of Education, Jilin University, Changchun, China
Correspondence:
[*]
Corresponding author. Hongwei Zhao, Department of computer science and technology, Jilin University, Changchun, China. E-mail: wangdanjl_dx@163.com.
Abstract: This paper designs a brand-new image retrieval method of mammary cancer based on convolution neural network. This method simulates VLAD layer in CNN network structure, designs a trainable universal VLAD layer-NET VLAD layer, reduces dimensions and optimizes VLAD descriptors, applies structure from motion algorithm to automatically label samples, and obtains the minimum loss function value by a new training program of weakly supervised ranking loss. Experiments show that this method has improved retrieval performance compared with similar retrieval methods and non-network structure retrieval methods.
Keywords: Convolutional neural network, VLAD, loss function, medical image retrieval
