Yong Ma

ORCID: 0000-0002-4139-9711
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Geological and Geophysical Studies
  • Hydrocarbon exploration and reservoir analysis
  • Video Analysis and Summarization
  • Handwritten Text Recognition Techniques
  • Context-Aware Activity Recognition Systems
  • Artificial Intelligence in Healthcare
  • Advanced Decision-Making Techniques
  • Image Processing Techniques and Applications
  • Robot Manipulation and Learning
  • Traffic Prediction and Management Techniques
  • Integrated Circuits and Semiconductor Failure Analysis
  • Distributed systems and fault tolerance
  • Robotic Mechanisms and Dynamics
  • Retinal Imaging and Analysis
  • Advanced Neural Network Applications
  • Human Pose and Action Recognition
  • High-Voltage Power Transmission Systems
  • Brain Tumor Detection and Classification
  • Advanced machining processes and optimization
  • Advanced Measurement and Metrology Techniques
  • Access Control and Trust
  • Iterative Learning Control Systems
  • Advanced Image Processing Techniques
  • Domain Adaptation and Few-Shot Learning
  • Spam and Phishing Detection

Beijing Institute of Technology
2024

Huazhong University of Science and Technology
2023

Wuhan University
2023

Nanjing University of Science and Technology
2022

Zhejiang Sci-Tech University
2021

South China University of Technology
2013

Shandong Institute of Automation
2007

Rongsheng Petrochemical (China)
2007

The identification and filtration of malicious texts in social media environments represent a significant technical challenge aimed at protecting users from online violence disinformation. This complexity stems the diversity innovativeness texts, which include unique expressions special sentence structures. Particularly, interrogative forms pose alignment challenges with traditional corpora due to existing methods' failure exploit text's deep global semantic representations. issue is...

10.1613/jair.1.16369 article EN cc-by Journal of Artificial Intelligence Research 2025-03-05

Computing the shortest path distance between two vertices in a road network is building block numerous applications. To do so efficiently, state-of-the-art proposals adopt tree decomposition process with heuristic strategies to build 2-hop label indexes. However, these indexes suffer from large space overheads caused by either imbalance or height. Independently of this, reinforcement learning has recently show impressive performance at sequential decision making spatial data management...

10.1109/icde55515.2023.00132 article EN 2022 IEEE 38th International Conference on Data Engineering (ICDE) 2023-04-01

The localization of texts in image/video is the first step a text processing system. Its effect will do great impact on following steps. Although many studies have been done algorithms, there not universally accepted performance evaluation method. In this paper we propose two sets metrics to evaluate algorithms different usage conditions. also consider distribution characteristics, and difficulties underlying task. Some experiments proposed are given.

10.1109/icdar.2007.4377072 article EN Proceedings of the International Conference on Document Analysis and Recognition 2007-09-01

Non-blind rotary motion deblurring (RMD) aims to recover the latent clear image from a blurred (RMB) image. The rotation center is crucial input parameter in non-blind RMD methods. Existing methods directly estimate RMB However they always suffer significant errors, and performance of limited. For assembled imaging systems, position remains fixed. Leveraging this prior knowledge, we propose geometric-based method for identification analyze its error range. Furthermore, construct system....

10.48550/arxiv.2408.04171 preprint EN arXiv (Cornell University) 2024-08-07

10.1109/cbs61689.2024.10860610 article EN 2018 IEEE International Conference on Cyborg and Bionic Systems (CBS) 2024-11-20

Zero-shot detection (ZSD) , the problem of object when training and test objects are disjoint, i.e. no examples target classes available.ZSD increasingly gains importance for large scale applications because collecting labeling sufficient data is extremely hard.In this paper, inspired from human cognitive experience, we propose a simple but effective Multi-label Context (MLC) framework to facilitate ability both seen unseen by mining contextual cues.We design multilabel classifier which...

10.18293/seke2022-012 article EN Proceedings/Proceedings of the ... International Conference on Software Engineering and Knowledge Engineering 2022-07-01

Due to the complicated theory of reliability analysis method and high requirement parameters statistics model calculation, most methods are not conductive engineering application. In order solve this problem, a simplified method, can be used in slope stability analysis. The technique is based on factor safety method. procedure applying studied. calculating standard deviation summarized. Engineering application shows that Taylor series feasible practical it provides new for

10.4028/www.scientific.net/amm.368-370.1769 article EN Applied Mechanics and Materials 2013-08-01

In order to carry out the construction scheme of oversize transmission overhead lines stringing successfully, paper has done research on by example Su Tong Transmission Overhead Lines so that a general can be formed and promoted use. As result, tension wires pulling ropes was obtained according formula calculation giving then achieved selection rope wires. Finally, developed strategy for Stringing. The will play key role in Oversize Long Span future.

10.12783/dtetr/amita2016/3598 article EN DEStech Transactions on Engineering and Technology Research 2016-11-17

The rotary motion deblurring is an inevitable procedure when the imaging seeker mounted in rotating missiles. Traditional methods suffer from ringing artifacts and noise, especially for large blur extents. To solve above problems, we propose a progressive framework consisting of coarse stage refinement stage. In first stage, design adaptive extents factor (BE factor) to balance noise suppression details reconstruction. And novel deconvolution model proposed based on BE factor. second...

10.1016/j.dt.2023.04.007 article EN cc-by-nc-nd Defence Technology 2023-04-15

This paper develops new methods for Trust Transfer in sparse environment based on indirectly feedback. relationship between the user and strange service provider was build adjusted automatically by other users. We present two techniques, first one is called initialization algorithm, which about to initialize trust provider, second named regulate algorithm dynamic period feedback mechanism. The simulation result demonstrate that this transfer method effective.

10.4028/www.scientific.net/amm.427-429.2268 article EN Applied Mechanics and Materials 2013-09-01
Coming Soon ...