Hong Deng

Hong Deng


Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 December 2019 in IEEE Access
Verified
Journal Article
Open access
YPYingqiong Peng
Yingqiong Peng
Muxin LiaoMuxin Liao
Muxin Liao
YSYuxia Song
Yuxia Song
ZLZhichao Liu
Zhichao Liu
HHHuojiao He
Huojiao He
YWYinglong Wang
Yinglong Wang

24 April 2017 in Food Analytical Methods
Verified
Journal Article
Closed access
PDF file
YPYingqiong Peng
Yingqiong Peng
WZWenjuan Zhang
Wenjuan Zhang
JCJuan Chang
Juan Chang
YHYaoping Huang
Yaoping Huang
LCLi Chen
Li Chen
YWYangping Wen
Yangping Wen

23 March 2020 in IEEE Access
Verified
Journal Article
Open access
YPYingqiong Peng
Yingqiong Peng
YSYuxia Song
Yuxia Song
WHWeiji Huang
Weiji Huang
HDHong Deng
Hong Deng
YWYinglong Wang
Yinglong Wang
JHJing Hua
Jing Hua
Cyber-Physical-Social Systems (CPSS) integrates cyber, physical and social spaces together, which makes our lives more convenient and intelligent by providing personalized service. In this paper, we will provide CPSS service for fine-grained recognition. Fine-grained visual recognition is a hot but challenging research in computer vision that aims to recognize object subcategories. The reason why it is challenging is that it extremely depends on the subtle discriminative features of local parts. Recently, some bilinear feature based methods were proposed, and the experimental results show state-of-the-art performance. However, most of them neglect the spatial relationships of part-region feature among multiple layers. In this paper, a novel approach of Self-layer and Cross-layer Bilinear Aggregation(SCBA) is proposed for fine-grained recognition. Firstly, a self-layer bilinear feature fusion module is proposed to model the spatial relationship of feature at the same layer. Secondly, we propose a cross-layer bilinear feature fusion module to capture the inter-layer interreaction of information to boost the ability of feature representation. In summary, the method we proposed not only can learn the correlations among different layers but the same layer, which makes it efficient and the experimental results show that it achieves state-of-the-art accuracy on three common fine-grained image datasets.

1 November 2018
Verified
Conference Paper
Closed access
YPYingqiong Peng
Yingqiong Peng
Muxin LiaoMuxin Liao
Muxin Liao
WHWeiji Huang
Weiji Huang
HDHong Deng
Hong Deng
LALing Ao
Ling Ao
JHJing Hua
Jing Hua
The recognition of fruit fly is an important part of the quarantine work, for which it is of great significance to establish the fruit fly feature automatic extraction classification system. Depending on the manual design of machine learning algorithm and the feature of the model classification, the existing fruit flies classification system needs to be improved in many ways, such as its complicated process, the excessive requirement for specialty, incapability of automatic feature extraction. Given these problems, the research on automatic feature extraction of classification model has been conducted, the convolutional neural network algorithm has been come up with. This algorithm can automatically extract feature classification training, try to find the solution to the problems of beforehand artificial designing and extracting in the existing fruit fly classification system, is helping to greatly improve the staff's efficiency. Experimental results show that the method can identify the Bactrocera dorsalis, Bactrocera cucurbitae, Bactrocera tau, Bactrocera scutellata, with the overall accuracy of 97.19 % , and have a good application prospect.