Toggle navigation
Home
About
About Journal
Historical Evolution
Indexed In
Awards
Reference Index
Editorial Board
Journal Online
Archive
Project Articles
Most Download Articles
Most Read Articles
Instruction
Contribution Column
Author Guidelines
Template
FAQ
Copyright Agreement
Expenses
Academic Integrity
Contact
Contact Us
Location Map
Subscription
Advertisement
中文
Journals
Publication Years
Keywords
Search within results
(((WANG Yihan[Author]) AND 1[Journal]) AND year[Order])
AND
OR
NOT
Title
Author
Institution
Keyword
Abstract
PACS
DOI
Please wait a minute...
For Selected:
Download Citations
EndNote
Ris
BibTeX
Toggle Thumbnails
Select
RefineDet based on subsection weighted loss function
XIAO Zhenyuan, WANG Yihan, LUO Jianqiao, XIONG Ying, LI Bailin
Journal of Computer Applications 2021, 41 (
7
): 1928-1932. DOI:
10.11772/j.issn.1001-9081.2020101615
Abstract
(
356
)
PDF
(1561KB)(
349
)
Knowledge map
Save
Concerning the poor performance of the Single-Shot Refinement Neural Network for Object Detection (RefineDet) of the object detection network when detecting small sample classes in inter-class imbalanced datasets, a Subsection Weighted Loss (SWLoss) function was proposed. Firstly, the inverse of the number of samples from different classes in each training batch was used as the heuristic inter-class sample balance factor to weight the different classes in the classification loss, thus strengthening the concern on the small sample class learning. After that, a multi-task balancing factor was introduced to weight classification loss and regression loss to reduce the difference between the learning rates of two tasks. At last, experiments were conducted on Pascal VOC2007 dataset and dot-matrix character dataset with large differences in the number of target class samples. The results demonstrate that compared to the original RefineDet, the SWLoss-based RefineDet clearly improves the detection precision of small sample classes, and has the mean Average Precision (mAP) on the two datasets increased by 1.01 and 9.86 percentage points, respectively; and compared to the RefineDet based on loss balance function and weighted pairwise loss, the SWLoss-based RefineDet has the mAP on the two datasets increased by 0.68, 4.73 and 0.49, 1.48 percentage points, respectively.
Reference
|
Related Articles
|
Metrics
Select
Provable radio frequency identification authentication protocol with scalability
SHI Zhicai, WANG Yihan, ZHANG Xiaomei, CHEN Shanshan, CHEN Jiwei
Journal of Computer Applications 2019, 39 (
3
): 774-778. DOI:
10.11772/j.issn.1001-9081.2018081648
Abstract
(
419
)
PDF
(817KB)(
266
)
Knowledge map
Save
The popular Radio Frequency IDentification (RFID) tags are some passive ones and they only have very limited computing and memory resources, which makes it difficult to solve the security, privacy and scalability problems of RFID authentication protocols. Based on Hash function, a security-provable lightweight authentication protocol was proposed. The protocol ensures the confidentiality and privacy of the sessions during the authentication process by Hashing and randomizing. Firstly, the identity of a tag was confirmed by its pseudonym and was preserved from leaking to any untrusted entity such as a reader. Secondly, only one Hashing computation was needed to confirm a tag's identity in the backend server, and the searching time to the tag's identity was limited to a constant by using the identifier to construct a Hash table. Finally, after each authentication, the secrecy and pseudonym of the tag were updated to ensure forward security of the protocol. It is proved that the proposed protocol satisfies scalability, forward security and anonymity demands and can prevent eavesdropping, tracing attack, replay attack and de-synchronization attack. The protocol only needs Hash function and pseudorandom generating operation for the tag, therefore it is very suitable to low-cost RFID systems.
Reference
|
Related Articles
|
Metrics