A5027230803- Photoacoustic and Ultrasonic Imaging
- Surgical Simulation and Training
- Sparse and Compressive Sensing Techniques
- Anatomy and Medical Technology
- Image and Signal Denoising Methods
- Image Processing Techniques and Applications
- Face and Expression Recognition
- Remote-Sensing Image Classification
- Remote Sensing in Agriculture
- Advanced Image Fusion Techniques
- Smart Agriculture and AI
- Remote Sensing and Land Use
- Topic Modeling
- Advanced Image Processing Techniques
- Rough Sets and Fuzzy Logic
- Optical Coherence Tomography Applications
- Image Enhancement Techniques
- Fuzzy Logic and Control Systems
- Soft Robotics and Applications
- Optical Systems and Laser Technology
- Remote Sensing and LiDAR Applications
- Augmented Reality Applications
Hunan Agricultural University
2024-2025
Xi'an Jiaotong University
2023
Accurately obtaining both the number and location of rice plants plays a critical role in agricultural applications, such as precision fertilization yield prediction. With rapid development deep learning, numerous models for plant counting have been proposed. However, many these contain large parameters, making them unsuitable deployment settings with limited computational resources. To address this challenge, we propose novel pruning method, Cosine Norm Fusion (CNF), lightweight feature...
Image denoising remains a fundamental challenge in digital image processing due to the inevitable presence of noise during acquisition and transmission. While existing filtering methods predominantly focus on local spatial information, they often overlook crucial structural information from other perspectives, such as manifold global structures. To address this limitation, we propose novel linear projection-based (LPNF) framework grounded projection learning theory. This innovatively learns...
BERT is a pre-trained language representation model that has received lot of attention for its impressive results in Natural Language Processing (NLP) tasks, such as sentiment analysis and text classification. This inspired many BERT-based models to be created improved on the original different ways. Examples these include RoBERTa, which improves BERTs pretraining; K-BERT, uses Knowledge Graphs improve domain-specific knowledge; others. In this paper, we will test with various datasets...
Although linear discriminant analysis (LDA)-based subspace learning has been widely applied to hyperspectral image (HSI) classification, the existing LDA-based methods exhibit several limitations: (1) They are often sensitive noise and demonstrate weak robustness; (2) these ignore local information inherent in data; (3) number of extracted features is restricted by classes. To address drawbacks, this paper proposes a novel joint sparse (JSLLDA) method integrating embedding regression...
This paper firstly illustrates knowledge representation of generalized modal syllogisms based on the structure quantification propositions, and proves validity syllogism ⧠ AM◇M-1 basis set theory logic. And then shows that other 37 valid can be deduced from AM◇M-1. Similarly, more derived it. The reason why these are reducible is necessary modality possible one ◇ able to define mutually, two Aristotelian quantifiers some no symmetric.
<p>Among the coded aperture compressed spectral imaging reconstruction algorithms, end-to-end method is a straightforward and effective method. However, presence of large number identical pixel points in measured image, mapped to different priori information, makes algorithm difficult solve accurately. In order enhance mapping characterization information process improve quality, we propose model optimization solving for encoding feature vector enhancement. Using wavelength position...
<p>Among the coded aperture compressed spectral imaging reconstruction algorithms, end-to-end method is a straightforward and effective method. However, presence of large number identical pixel points in measured image, mapped to different priori information, makes algorithm difficult solve accurately. In order enhance mapping characterization information process improve quality, we propose model optimization solving for encoding feature vector enhancement. Using wavelength position...
Compressive spectral imaging (CSI) is a snapshot technique that rapidly captures the information of target in single exposure and effectively reconstructs high-spectral data using reconstruction algorithms. However, due to presence large number identical pixels measured image, which map different prior information, existing algorithms struggle establish an accurate pixel separation representation model. In order improve effect between enhance capability image pixels, we propose compressed...