A5048790634- Particle physics theoretical and experimental studies
- Quantum Chromodynamics and Particle Interactions
- High-Energy Particle Collisions Research
- Computational Physics and Python Applications
- Black Holes and Theoretical Physics
- Advanced Steganography and Watermarking Techniques
- Chaos-based Image/Signal Encryption
- Digital Media Forensic Detection
- Neutrino Physics Research
- Energy Efficient Wireless Sensor Networks
- Stochastic processes and statistical mechanics
- Security in Wireless Sensor Networks
- advanced mathematical theories
- Time Series Analysis and Forecasting
- Mobile Ad Hoc Networks
- Particle Accelerators and Free-Electron Lasers
- Artificial Intelligence in Law
- Network Security and Intrusion Detection
- Speech Recognition and Synthesis
- Advanced Data Compression Techniques
- Music and Audio Processing
- Smart Grid Energy Management
- AI in cancer detection
- Medical Image Segmentation Techniques
- Particle Detector Development and Performance
Liaoning Normal University
2023-2025
Zhengzhou University
2023-2025
Macao Polytechnic University
2022-2025
China Agricultural University
2024-2025
Southern University of Science and Technology
2025
Jilin Medical University
2024
Jilin University
2024
University of Shanghai for Science and Technology
2023
North University of China
2021
University of Science and Technology of China
2007-2010
The smart healthcare system plays a vital role in modern healthcare, facilitating the exchange of Electronic Patient Records (EPR) and improving medical care. Nonetheless, safeguarding inherent security confidentiality EPR data persists as formidable challenge, demanding rigorous attention innovative solutions. Digital watermarking safeguards genuineness integrity digital images is widely employed. In imaging, guaranteeing patient vital. this paper, we propose an...
LLM-as-a-Judge, which generates chain-of-thought (CoT) judgments, has become a widely adopted auto-evaluation method. However, its reliability is compromised by the CoT reasoning's inability to capture comprehensive and deeper details, often leading incomplete outcomes. Existing methods mainly rely on majority voting or criteria expansion, insufficient address limitation in CoT. We propose Crowd-based Comparative Evaluation, introduces additional crowd responses compare with candidate...
In the wake of accelerated advancement Internet Things(IoT) technology, a significant volume multimedia information is emerging as dominant component IoT applications. However, this also led to big security risks. Therefore, secure and reliable image encryption scheme based on non-degenerate hyper chaos dynamic bidirectional permutation proposed for characteristic digital images which primary carrier in communication environment. The uses discrete generate pseudo-random sequences applies...
Phyllosticta fragaricola-induced angular leaf spot causes substantial economic losses in global strawberry production, necessitating advanced severity assessment methods. This study proposed a dual-phase grading framework integrating deep learning and computer vision. The enhanced You Only Look Once version 11 (YOLOv11) architecture incorporated Content-Aware ReAssembly of FEatures (CARAFE) module for improved feature upsampling squeeze-and-excitation (SE) attention mechanism channel-wise...
Learning from datasets without interaction with environments (Offline Learning) is an essential step to apply Reinforcement (RL) algorithms in real-world scenarios. However, compared the single-agent counterpart, offline multi-agent RL introduces more agents larger state and action space, which challenging but attracts little attention. We demonstrate current are ineffective systems due accumulated extrapolation error. In this paper, we propose a novel algorithm, named Implicit Constraint...
The automated collection of plant phenotypic information has become a trend in breeding and smart agriculture. Four YOLOv8-based models were used to segment mature soybean plants placed simple background laboratory environment, identify pods, distinguish the number soybeans each pod, obtain phenotypes. YOLOv8-Repvit model yielded most optimal recognition results, with an R2 coefficient value 0.96 for both pods beans, RMSE values 2.89 6.90, respectively. Moreover, novel algorithm was devised...
In this paper we propose a blind dual watermarking scheme using Set Partitioning in Hierarchical Trees (SPIHT) and Stucki Kernel halftone technique for the tamper detection image self-recovery. The watermark consists of authentication bits tampering area location recovery restoration. We generate two to ensure high-quality tampered image. primary bit is generated by SPIHT encoding, secondary technique. Then based on bits. Before embedding watermark, shuffle through Arnold cat mapping...
Sensor nodes can be compromised by attackers to launch various attacks, one of which is false data injection attack The attacker injects large amount bogus through the node. This will lead network resources exhaustion. We propose a novel contact-based approach (CTrace) traceback origins. In CTrace, each node maintains information its neighbors in proximity radius R hops. at hops away called contact. With certain probability, marks identification into an arriving packet. By collecting few...
In this paper, we propose a novel Comprehensive-Monitoring-Motivated Multi-model Watermarking (C3MW) Scheme for tampering region localization and self-recovery 4K images. To generate the (C3M) Watermark, Multi-Model Authentication Bits Generation (MMAG) method Adaptive Block Significance-Based Recovery (ASRG) method, are proposed. The MMAG aims to monitor various bit-plane information detecting possible in more comprehensive manner. When performing image detection, once one of multiple...
With significant efforts in recent studies, LLM-as-a-Judge has become a cost-effective alternative to human evaluation for assessing the text generation quality wide range of tasks. However, there still remains reliability gap between and evaluation. One important reason is lack guided oracles process. Motivated by role reference pervasively used classic evaluation, we introduce RevisEval, novel paradigm via response-adapted references. RevisEval driven key observation that an ideal should...
False data injection is a big threat to sensor networks. A traceback mechanism based on probabilistic packet marking proposed in this paper. The performance of the basic method thoroughly studied. There defect that upstream nodes' marks are collected by sink with low probability. To solve problem, two different improved methods-EPPM&EPNM-are proposed. Every node's approximately equal probability EPPM. can locate every node collecting packets number EPNM.
Speech is the most natural information carrier for human beings, and it likely to become main way of human-computer interaction in future. This paper presents an isolated score word recognition method using Mel-scale Frequency Cepstral Coefficients (MFCC) Dynamic Time Warping (DTW). The processing stage speech signal basic system, analyze convert into feature parameters. An endpoints detection proposed joint adjustment short-term energy zero-crossing rate. It can better detect endpoints,...
This paper proposes a halftoning-based fragile watermarking approach for digital image tamper detection and self-recovery. The Set Partitioning in Hierarchical Trees (SPIHT) algorithm halftoning technique are employed to generate the primary recovery bits secondary bits, respectively. On basis of that, authentication generated. Arnold Cat Map diagonal mapping then applied further improve accuracy ensure quality experimental results have been conducted demonstrate superiorities proposed...
Cover lossless robust watermarking technology is one of the frontier hotspots in field information hiding. It can fully restore original image without external attacks, while when there an attack, it extract watermarks for secure verification. Most countermeasures proposed literature are to geometric but most schemes less imperceptible, mainly due fixed location embeddings. In this paper, we propose a Zernike moments-based and reversible scheme. embeds low-order moments keep their rotational...
This paper proposes a dual-tampering detection method for 1080p images, then further recovering its forged content with the recovery information, which is generated by low-frequency of second-order wavelet. First, image divided into non-overlapping blocks. To improve quality recovered image, we employ centrosymmetric mapping technique that ensures security bits. MD5 used to generate first part authentication key based on most significant bit-plane block. SHAI second information. By embedding...