Zhiyu Jiang

Prof. Zhiyu Jiang

Northwestern Polytechnical University, 127 West Youyi Road, Xi'an Shaanxi, 710072, P.R. China
orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

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

Representative Publications

View All Publications
1 November 2015 in Neurocomputing
Verified
Journal Article
Closed access
YYYuan Yuan
Yuan Yuan
ZJZhiyu Jiang
Zhiyu Jiang
Qi WangQi Wang
Qi Wang

28 May 2025 in Pattern Recognition
Verified
Journal Article
Closed access
ZJZhiyu Jiang
Zhiyu Jiang
JLJianing Li
Jianing Li
SXShijie Xu
Shijie Xu
ZLZhuozhao Liu
Zhuozhao Liu
DMDandan Ma
Dandan Ma
YYYuan Yuan
Yuan Yuan

1 February 2024 in JAMA Ophthalmology
Verified
Journal Article
Open access
PDF file
YYYahan Yang
Yahan Yang
HLHuipeng Liao
Huipeng Liao
LZLanqin Zhao
Lanqin Zhao
XWXun Wang
Xun Wang
XYXiaoWei Yang
XiaoWei Yang
HLHaotian Lin
Haotian Lin
Importance: China has experienced both rapid urbanization and major increases in myopia prevalence. Previous studies suggest that green space exposure reduces the risk of myopia, but the association between myopia risk and specific geometry and distribution characteristics of green space has yet to be explored. These must be understood to craft effective interventions to reduce myopia.Objective: To evaluate the associations between myopia and specific green space morphology using novel quantitative data from high-resolution satellite imaging.Design, Setting, and Participants: This prospective cohort study included students grades 1 to 4 (aged 6 to 9 years) in Shenzhen, China. Baseline data were collected in 2016-2017, and students were followed up in 2018-2019. Data were analyzed from September 2020 to January 2022.Exposures: Eight landscape metrics were calculated using land cover data from high-resolution Gaofen-2 satellite images to measure area, aggregation, and shape of green space.Main Outcome and Measures: The 2-year cumulative change in myopia prevalence at each school and incidence of myopia at the student level after 2 years were calculated as main outcomes. The associations between landscape metrics and school myopia were assessed, controlling for geographical, demographic, and socioeconomic factors. Principal component analyses were performed to further assess the joint effect of landscape metrics at the school and individual level.Results: A total of 138 735 students were assessed at baseline. Higher proportion, aggregation, and better connectivity of green space were correlated with slower increases in myopia prevalence. In the principal component regression, a 1-unit increase in the myopia-related green space morphology index (the first principal component) was negatively associated with a 1.7% (95% CI, −2.7 to −0.6) decrease in myopia prevalence change at the school level (P = .002). At the individual level, a 1-unit increase in myopia-related green space morphology index was associated with a 9.8% (95% CI, 4.1 to 15.1) reduction in the risk of incident myopia (P < .001), and the association remained after further adjustment for outdoor time, screen time, reading time, and parental myopia (adjusted odds ratio, 0.88; 95% CI, 0.80 to 0.97; P = .009).Conclusions and Relevance: Structure of green space was associated with a decreased relative risk of myopia, which may provide guidance for construction and renovation of schools. Since risk estimates only indicate correlations rather than causation, further interventional studies are needed to assess the effect on school myopia of urban planning and environmental designs, especially size and aggregation metrics of green space, on school myopia.

1 January 2022 in Ophthalmology
Verified
Journal Article
Open access
YYYahan Yang
Yahan Yang
WCWenben Chen
Wenben Chen
AXAndi Xu
Andi Xu
LZLanqin Zhao
Lanqin Zhao
XDXiaohu Ding
Xiaohu Ding
HLHaotian Lin
Haotian Lin

1 April 2018 in Neurocomputing
Verified
Journal Article
Closed access
ZJZhiyu Jiang
Zhiyu Jiang
YYYuan Yuan
Yuan Yuan
Qi WangQi Wang
Qi Wang

6 June 2021
Verified
Conference Paper
Closed access
ZCZhinan Cai
Zhinan Cai
ZJZhiyu Jiang
Zhiyu Jiang
YYYuan Yuan
Yuan Yuan
Change detection for remote sensing images is widely applied for urban change detection, disaster assessment and other fields. However, most of the existing CNN-based change detection methods still suffer from the problem of inadequate pseudo-changes suppression and insufficient feature representation. In this work, an unsupervised change detection method based on Task-related Self-supervised Learning Change Detection network with smooth mechanism(TSLCD) is proposed to eliminate it. The main contributions include: (1) the task-related self-supervised learning module is introduced to extract spatial features more effectively. (2) a hard-sample-mining loss function is applied to pay more attention to the hard-to-classify samples. (3) a smooth mechanism is utilized to remove some of pseudo-changes and noise. Experiments on four remote sensing change detection datasets reveal that the proposed TSLCD method achieves the state-of-the-art for change detection task.