Open Access Open Access  Restricted Access Subscription Access

Prompt Image Search with Deep Convolutional Neural Network via Efficient Hashing Code and Addictive Latent Semantic Layer

Jun-yi Li,
Jian-hua Li,

Abstract


As we know that the nearest neighbor search is a good and effective method for good-sized image search. This paper indicates a vision learning framework to generate compact binary hash codes for quick vision search after knowing the recent benefits of convolution neural networks (CNN). Our concept is that binary codes can be obtained using a hidden layer to present some latent concepts dominating the class labels with usable data labels. CNN also can be used to learn image representations. Binary code learning is required for other supervised methods. However, our method is effective in obtaining hash codes and image representations and we use pretrained model from googlenet for incremental learning so it is suitable for good-sized dataset. It is demonstrated in our experiment that this method is better than some most advanced hashing algorithms in MINIST, NUS-WIDE and CIFAR-10 dataset. The scalability and efficiency still needs to be further investigated in a good-sized dataset.


Citation Format:
Jun-yi Li, Jian-hua Li, "Prompt Image Search with Deep Convolutional Neural Network via Efficient Hashing Code and Addictive Latent Semantic Layer," Journal of Internet Technology, vol. 19, no. 3 , pp. 949-957, May. 2018.

Full Text:

PDF

Refbacks

  • There are currently no refbacks.





Published by Executive Committee, Taiwan Academic Network, Ministry of Education, Taipei, Taiwan, R.O.C
JIT Editorial Office, Library and Information Center, National Dong Hwa University
No. 1, Sec. 2, Da Hsueh Rd. Shoufeng, Hualien 97401, Taiwan, R.O.C.
Tel: +886-3-931-7017  E-mail: jit.editorial@gmail.com