A5101427624- Financial Distress and Bankruptcy Prediction
- Imbalanced Data Classification Techniques
- Biometric Identification and Security
- Rough Sets and Fuzzy Logic
- Robotics and Sensor-Based Localization
- Advanced Computational Techniques and Applications
- Face recognition and analysis
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Video Surveillance and Tracking Methods
- AI-based Problem Solving and Planning
- Maritime and Coastal Archaeology
- Forensic Fingerprint Detection Methods
- Water Resources and Management
- Geoscience and Mining Technology
- Physical Unclonable Functions (PUFs) and Hardware Security
- Image and Object Detection Techniques
- Embedded Systems and FPGA Design
- Thermography and Photoacoustic Techniques
- Image and Video Stabilization
- Management and Optimization Techniques
- User Authentication and Security Systems
- Advanced Computational Techniques in Science and Engineering
- Machine Learning and Data Classification
Nanjing Institute of Railway Technology
2025
Zhejiang Water Conservancy and Hydropower Survey and Design Institute
2023
Zhejiang Gongshang University
2012-2023
Qingdao University of Technology
2013-2022
Qingdao University of Science and Technology
2022
North China University of Technology
2020-2021
University of Science and Technology of China
2017
Gyeongju University
2015
Zhejiang Normal University
2007-2015
Dongguk University
2015
Ship instance segmentation technologies enable the identification of ship targets and their contours, serving as an auxiliary tool for monitoring, tracking, providing critical support maritime port safety management. However, due to different shapes sizes ships, well complexity fluctuation lighting weather, existing approaches frequently struggle accomplish correct contour segmentation. To address this issue, paper introduces Contour, a real-time method instances based on contours that...
Abstract The existing iris recognition methods offer excellent performance for known classes, but they do not consider the rejection of unknown classes. It is important to reject an object class a reliable system. This study proposes open‐set based on deep learning. In method, by training network, extracted features are clustered near feature centre each kind image. Then, authors build open‐class outlier network (OCFON) containing distance features, which maps new space and classifies them....
An ideal model for credit risk assessment is supposed to select important features and process imbalanced data sets in an effective manner. This paper proposes integrated method that combines B&B (branch bound)-based hybrid feature selection (BBHFS) with the imbalanceoriented multiple-classifier ensemble (IOMCE) uses support vector machine (SVM) multiple discriminant analysis (MDA) as base predictor. BBHFS a integrates t-test k-fold crossvalidation search satisfactory subset. The IOMCE...
Aiming at the excessive dependence of traditional path planning methods on map information and lack self-learning self-adaptive capabilities, a method for robot based deep reinforcement learning is proposed. The paper takes lidar data Gazebo simulation environments built ROS platform as input. Learn direct action control from environment through end-to-end learning, adopt neural network to fit value-based non-model time difference Q algorithm, reasonably design models, number state spaces,...
The polynomial regression algorithm can fit almost all data sequences by adding high-order terms of independent variables, but on long sequences, it only reflect the development trends and predicted value is not accurate enough. RNN has great advantages in predicting fitting time sequence data, especially Long Short-Term Memory (LSTM) network good performance sequence. This paper respectively adopts LSTM analysis predict effect epidemic prevention control policies various countries to...
The continuously widening design productivity gap in the past few decades gives high incentives to counterfeiting ICs. Existing intellectual property (IP) protection schemes demand overheads, and some techniques like watermarking do not facilitate tracing of illegal users. In this letter, we propose a novel fingerprinting method based on post-processing circuit partitions. We evaluate our ISPD98 benchmark suite. experimental results demonstrate effectiveness proposal. is distinct because...
Small hydropower is an important infrastructure for people's livelihoods and clean renewable energy, but there are still many weak links in the planning, design, construction, operation, management of small some regions. Kaihua County located at source Qianjiang River, rational development utilization its water resources great significance protecting ecological environment River achieving strategy goal "ecological county characteristic county" County. To alleviate reduction caused by...
Using Martin Meter from Japan to measure 110 young women body. SPSS software for correlation analysis of measurement data, then selecting width, thickness and weight as elements used establishing leg girth regression models. Adopting backward method choose variable further, Multiple Linear model, Multivariate Quadratic Cubic model predict the each part which related clothing. After testing comparing effect choosing out best fitting modle part. The validity modles were tested by randomly 20...