Song Luo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Song Luo has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Ning Hu
Kai Feng Culture And Arts College, Kaifeng 475000, China
1 shared publication - Qingni Shen
Peking University
1 shared publication - Zhikun Wang
Chongqing University of Technology,School of Computer Science and Engineering,China
1 shared publication Zhong Chen
Peking University
1 shared publication- Zhuang Chen
College of Electronic and Information Engineering, Southwest University, Chongqing, China
1 shared publication
Latest Publications
- 29 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 433-436
Build a real-time building segmentation network
Recently, most existing building segmentation networks have high memory usage and computational resource. And a large number of methods of the lightweight model have problems, such as the sharp decrease in segmentation performance due to parameter reduction, limited arithmetic power and real-time. Therefore, in this paper, we propose a new lightweight network based on the SGUnet segmentation network. Our network is a hourglass-like-depth-separable-convolution-based sgbunet segmentation network, pruning the sgnet and using residual connections. The core idea is that, shallow features are more important in lightweight networks. According to this point of view, increasing the proportion of features in the shallow part of the model successfully improved the accuracy of the model by 1%. Finally, the Cheap operation module is reused to further reduce the amount of network parameters. In the end, the parameter amount of the network is reduced from the original 260k to 70k, which greatly reduces the amount of model parameters.The sgbunet segmentation network have a reletive improvement, achieving 97.329 and 80.823 for Accuracy and Miou respectively in the building segmentation.
- 29 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 433-436
Build a real-time building segmentation network
Recently, most existing building segmentation networks have high memory usage and computational resource. And a large number of methods of the lightweight model have problems, such as the sharp decrease in segmentation performance due to parameter reduction, limited arithmetic power and real-time. Therefore, in this paper, we propose a new lightweight network based on the SGUnet segmentation network. Our network is a hourglass-like-depth-separable-convolution-based sgbunet segmentation network, pruning the sgnet and using residual connections. The core idea is that, shallow features are more important in lightweight networks. According to this point of view, increasing the proportion of features in the shallow part of the model successfully improved the accuracy of the model by 1%. Finally, the Cheap operation module is reused to further reduce the amount of network parameters. In the end, the parameter amount of the network is reduced from the original 260k to 70k, which greatly reduces the amount of model parameters.The sgbunet segmentation network have a reletive improvement, achieving 97.329 and 80.823 for Accuracy and Miou respectively in the building segmentation.
- Open Access
- 31 December 2013
- Published by Wiley in Journal of Applied Mathematics
- Vol. 2013, 1-9
On the Anonymity of Identity-Based Encryption
Anonymity of identity-based encryption (IBE) means that given a ciphertext, one cannot distinguish the target identity from a random identity. In this paper, we thoroughly discuss the anonymity of IBE systems. We found that the current definition of anonymity is obscure to describe some IBE systems, such as Gentry IBE system. Furthermore, current definition cannot express the degree of anonymity. So we divide the degree of anonymity into weak anonymity and strong anonymity based on indistinguishability between different games. For weakly anonymous IBE systems, the target identity in a ciphertext cannot be distinguished from a random identity. For strongly anonymous IBE systems, the whole ciphertext cannot be distinguished from a random tuple. We also discuss the type of anonymity and divide it into two types. Type 1 means that a random tuple can be seen as a valid ciphertext, while type 2 cannot. Based on our new definitions, we show that three famous IBE systems, Gentry IBE system, Boyen-Waters IBE system, and Lewko IBE system, have strong but different types of anonymity.
- 31 December 2013
- Published by Wiley in Journal of Applied Mathematics
- Vol. 2013, 1-9
Open AccessOn the Anonymity of Identity-Based Encryption
Anonymity of identity-based encryption (IBE) means that given a ciphertext, one cannot distinguish the target identity from a random identity. In this paper, we thoroughly discuss the anonymity of IBE systems. We found that the current definition of anonymity is obscure to describe some IBE systems, such as Gentry IBE system. Furthermore, current definition cannot express the degree of anonymity. So we divide the degree of anonymity into weak anonymity and strong anonymity based on indistinguishability between different games. For weakly anonymous IBE systems, the target identity in a ciphertext cannot be distinguished from a random identity. For strongly anonymous IBE systems, the whole ciphertext cannot be distinguished from a random tuple. We also discuss the type of anonymity and divide it into two types. Type 1 means that a random tuple can be seen as a valid ciphertext, while type 2 cannot. Based on our new definitions, we show that three famous IBE systems, Gentry IBE system, Boyen-Waters IBE system, and Lewko IBE system, have strong but different types of anonymity.
- 1 January 2012
- Published by Springer Nature
- p. 109-126
Fully Secure Unidirectional Identity-Based Proxy Re-encryption
- 1 January 2012
- Published by Springer Nature
- p. 109-126
Fully Secure Unidirectional Identity-Based Proxy Re-encryption