A5064144631- Spectroscopy and Chemometric Analyses
- Smart Agriculture and AI
- Complex Systems and Time Series Analysis
- Market Dynamics and Volatility
- Financial Markets and Investment Strategies
- Remote Sensing in Agriculture
- Bamboo properties and applications
- Genomics and Phylogenetic Studies
- Plant Pathogens and Fungal Diseases
- Water Quality Monitoring and Analysis
- Horticultural and Viticultural Research
- Risk and Portfolio Optimization
- Stock Market Forecasting Methods
- Plant Disease Management Techniques
- Financial Risk and Volatility Modeling
- Soil Geostatistics and Mapping
- Spatial and Panel Data Analysis
- Domain Adaptation and Few-Shot Learning
- Monetary Policy and Economic Impact
- Housing Market and Economics
- Advanced Neural Network Applications
- Wood and Agarwood Research
- Evaluation Methods in Various Fields
- Banking stability, regulation, efficiency
- Geochemistry and Geologic Mapping
Northeastern University
2010-2025
Anhui Agricultural University
2010-2024
Ministry of Agriculture and Rural Affairs
2022-2024
Shenyang University
2018
Universidad del Noreste
2014
Los Alamos National Security (United States)
2004
Classification of healthy and diseased wheat heads in a rapid non-destructive manner for the early diagnosis Fusarium head blight disease research is difficult. Our work applies deep neural network classification algorithm to pixels hyperspectral image accurately discern area. The spectra manually selected region interest are preprocessed via mean removal eliminate interference, due time interval environment. generalization model considered, two improvements made framework. First, pixel data...
The application of visual-near-infrared and shortwave-infrared (VNIR-SWIR) diffuse reflectance spectroscopy for soil properties analysis is increasingly gaining popularity due to its rapid, cost-effective, non-destructive nature. In particular, deep learning models have been found perform exceptionally well large spectra libraries. This study proposes a novel approach enhance the that involves converting one-dimensional into two-dimensional (2D) spectral images. We investigated several...
The application of visible near-infrared (VIS-NIR) analysis technology to quantify the nutrients in soil has been widely recognized. It is important improve performance regression models that can predict soil-available potassium concentration. This study collected samples from southern Anhui, China, and concentrated on modelling methods by using 29 pretreatment methods. results show a combination three methods, Savitzky–Golay, standard normal variate, dislodge tendency, exhibited better...
Purpose Previous research has predominantly concentrated on examining risk spillovers through single-layer networks, neglecting the multi-related and multilayer network characteristics of economic system. This study constructs connectedness including return, volatility extreme layers, to systematically analyze across Chinese industries at system industry levels. Design/methodology/approach studies have constructed networks using Diebold Yilmaz’s (2012) approach or time-varying parameter...
Abstract Although the classification method based on deep neural network has achieved excellent results in tasks, it is difficult to apply real-time scenarios because of high memory footprints and prohibitive inference times. Compared unstructured pruning, structured pruning techniques can reduce computation cost model runtime more effectively, but inevitably reduces precision model. Traditional methods use fine tuning restore damage performance. However, there still a large gap between...
Orphan genes are associated with regulatory patterns, but experimental methods for identifying orphan both time-consuming and expensive. Designing an accurate robust classification model to detect non-orphan in unbalanced distribution datasets poses a particularly huge challenge. Synthetic minority over-sampling algorithms (SMOTE) selected preliminary step deal gene datasets. To identify balanced Arabidopsis thaliana datasets, SMOTE were then combined traditional advanced ensemble classified...
Precise pear detection and recognition is an essential step toward modernizing orchard management. However, due to the ubiquitous occlusion in orchards various locations of image acquisition, pears acquired images may be quite small occluded, causing high false object loss rate. In this paper, a multi-scale collaborative perception network YOLOv5s-FP (Fusion Perception) was proposed for detection, which coupled local global features. Specifically, dataset with proportion occluded proposed,...
The “Dangshan” pear woolliness response is a physiological disease that mostly occurs in the growth process. appearance of not obvious, and it difficult to detect with naked eye. Therefore, finding way quickly nondestructively identify great significance. In this paper, near-infrared spectral (NIR) data samples were collected at 900–1700 nm reflectance spectra using handheld miniature NIR spectrometer, modelled analysed random forest (RF), support vector machine (SVM) boosting algorithms...
The "Dangshan" pear woolliness response is a physiological disease that causes large losses for fruit farmers and nutrient inadequacies.The cause of this predominantly shortage boron calcium in the water loss from pear. This paper used fusion near-infrared Spectroscopy (NIRS) Computer Vision Technology (CVS) to detect pears. employs merging NIRS features image detection disease. Near-infrared reflects information on organic matter containing hydrogen groups other components various...
Accurate diagnosis of pear tree nutrient deficiency symptoms is vital for the timely adoption fertilization and treatment. This study proposes a novel method on fused feature multi-head attention recording network with image depth shallow fusion diagnosing in leaves. First, features nutrient-deficient leaf images are extracted using manual extraction methods, by deep model. Second, serial fusion. In addition, trained three classification algorithms, F-Net, FC-Net, FA-Net, proposed this...