- Spectroscopy and Chemometric Analyses
- Remote Sensing in Agriculture
- Smart Agriculture and AI
- Advanced Chemical Sensor Technologies
- Leaf Properties and Growth Measurement
- Spectroscopy Techniques in Biomedical and Chemical Research
- Plant Pathogenic Bacteria Studies
- Pesticide and Herbicide Environmental Studies
- Weed Control and Herbicide Applications
- Aviation Industry Analysis and Trends
- Fermentation and Sensory Analysis
- Advanced Measurement and Detection Methods
- Remote Sensing and Land Use
- Photosynthetic Processes and Mechanisms
- Evaluation Methods in Various Fields
- Gas Sensing Nanomaterials and Sensors
- Transportation and Mobility Innovations
- Industrial Vision Systems and Defect Detection
- Greenhouse Technology and Climate Control
- Terahertz technology and applications
- Crop Yield and Soil Fertility
- Color Science and Applications
- Elevator Systems and Control
- Plant Pathogens and Fungal Diseases
- Biosensors and Analytical Detection
Purdue University West Lafayette
2022-2024
Zhejiang University
2018-2022
Ministry of Agriculture and Rural Affairs
2020-2022
Zhengzhou Normal University
2022
State Key Laboratory of Modern Optical Instruments
2021
University of Sheffield
2021
Hangzhou Academy of Agricultural Sciences
2020
Institute of Spectroscopy
2020
China University of Geosciences
2017-2018
Because bacterial blight (BB) disease seriously affects the yield and quality of rice, breeding BB resistant rice is an important priority for plant breeders but process time-consuming. The feasibility using terahertz imaging technology near-infrared hyperspectral to identify seeds has therefore been studied. two-dimensional (2D) spectral images one-dimensional (1D) spectra provided by both methods were used build discriminant models based on a deep learning method, convolutional neural...
The object detection method based on deep learning convolutional neural network (CNN) significantly improves the performance of wheat head images obtained from near ground. Nevertheless, for different stages, high density, and overlaps captured by aerial-scale unmanned aerial vehicle (UAV), existing learning-based methods often have poor effects. Since receptive field CNN is usually small, it not conducive to capture global features. visual Transformer can information an image; hence we...
Near-infrared (874–1734 nm) hyperspectral imaging technology combined with chemometrics was used to identify parental and hybrid okra seeds. A total of 1740 seeds three different varieties, which contained the male parent xiaolusi, female xianzhi, seed penzai, were collected, all samples randomly divided into calibration set prediction in a ratio 2:1. Principal component analysis (PCA) applied explore separability based on spectral characteristics Fourteen 86 characteristic wavelengths...
Bacterial blight poses a threat to rice production and food security, which can be controlled through large-scale breeding efforts toward resistant cultivars. Unmanned aerial vehicle (UAV) remote sensing provides an alternative means for the infield phenotype evaluation of crop disease resistance relatively time-consuming laborious traditional methods. However, quality data acquired by UAV affected several factors such as weather, growth period, geographical location, limit their utility...
Diseases caused by invasive pathogens are an increasing threat to forest health, and early accurate disease detection is essential for timely precision management. The recent technological advancements in spectral imaging artificial intelligence have opened up new possibilities plant both crops trees. In this study, Dutch elm (DED; Ophiostoma novo-ulmi,) American (Ulmus americana) was used as example pathosystem evaluate the accuracy of two in-house developed high-precision portable hyper-...
The feasibility of using the fourier transform infrared (FTIR) spectroscopic technique with a stacked sparse auto-encoder (SSAE) to identify orchid varieties was studied. Spectral data 13 orchids covering spectral range 4000–550 cm−1 were acquired establish discriminant models and select optimal variables. K nearest neighbors (KNN), support vector machine (SVM), SSAE built full spectra. model performed better than KNN SVM obtained classification accuracy 99.4% in calibration set 97.9%...
Rice seed vigor plays a significant role in determining the quality and quantity of rice production. Thus, quick non-destructive identification is not only beneficial to fully obtain state seeds but also intelligent development agriculture by instant monitoring. herein, near-infrared hyperspectral imaging technology, as an information acquisition tool, was introduced combined with deep learning algorithm identify vigor. Both spectral images average spectra were sent discriminant models...
Hyperspectral imaging technique combined with machine learning is a powerful tool for the evaluation of disease phenotype in rice disease-resistant breeding. However, current studies are almost carried out lab environment, which difficult to apply field environment. In this paper, we used visible/near-infrared hyperspectral images analysis severity bacterial blight (BB) and proposed novel index construction strategy (NDSCI) application. A designed long short-term memory network attention...
Spirulina platensis can synthesize a large amount of phycocyanin, which had been developed as health food. At the same time, absorb nitrogen and phosphorus in wastewater provide for its own growth. Here, we studied optimal supply production process. For first 405 nm portable Raman spectrometer was used to estimate phycocyanin content real-time industrial applications. We obtained three characteristic peaks through density functional theory combined with home-built spectrometer, were 1272,...
Remote sensing coupled with hyperspectral technology has become increasingly popular to investigate plant traits, showcasing its advantages in studying growth, health, and productivity. The quality of the collected images is crucial for subsequent data analysis phenotyping studies. However, diurnal variations spectral characteristics introduce more variance canopy reflectance spectra, raising cost analyses compromising performance trait estimation models. In this study, a fixed gantry...
Abstract Developing herbicide resistant cultivars is one of the effective methods to solve safety problem caused by use herbicide. In this study, hyperspectral image was used develop more robust leaf chlorophyll content (LCC) prediction model based on different datasets finally analyze response LCC glyphosate-stress. Chlorophyll a fluorescence (ChlF) dynamically monitor photosynthetic physiological transgenic glyphosate-resistant and wild glyphosate–sensitive maize seedlings, applying...
Accurate estimation of leaf pose in the field environment provides a pivotal foundation for precision agriculture applications ranging from automation proximal sensing, sampling, and harvesting. However, current methods have limited performance because they predominantly depend on preselected features leaves, making them suboptimal leaves that lack characteristics. This study presents novel deep learning approach to estimate with high robustness. The method utilized ResNet-based model...
Camera-based tactile sensors can provide high-density surface geometry and force information for robots in the interaction process with target. However, most existing methods cannot achieve accurate reconstruction high efficiency, impeding applications robots. To address these problems, we propose an efficient two-shot photometric stereo method based on symmetric color LED distribution. Specifically, sensing response curve of CMOS channels, design orthogonal red blue LEDs as illumination to...
Currently, the presence of genetically modified (GM) organisms in agro-food markets is strictly regulated by enacted legislation worldwide. It essential to ensure traceability these transgenic products for food safety, consumer choice, environmental monitoring, market integrity, and scientific research. However, detecting existence GM involves a combination complex, time-consuming, labor-intensive techniques requiring high-level professional skills. In this paper, concise rapid pipeline...
Glyphosate is a widely used nonselective herbicide. Probing the glyphosate tolerance mechanism necessary for screening and development of resistant cultivars. In this study, hyperspectral image was to develop more robust leaf chlorophyll content (LCC) prediction model based on different datasets finally analyze response LCC glyphosate-stress. Chlorophyll fluorescence (ChlF) dynamically monitor photosynthetic physiological transgenic glyphosate-resistant wild glyphosate-sensitive maize...
Phosphorus (P) is a vital macronutrient for building up essential biomolecules in plants, and its accurate quantification can guide effective crop management increase growers’ profit. Traditional chemical reaction-based methods measuring P levels plants are destructive complex. Hyperspectral imaging offers real-time, non-destructive avenue assessing nutrient status. While these images rich both spatial spectral information, limitations current devices analytical algorithms have led most...
As one of the largest supplied grain crops, corn plants often require a significant amount nitrogen fertilizer for optimal yield. However, excessive usage can lead to adverse environmental consequences, especially nearby hydrological network. To precisely manage application, accurate measurement crop deficiency is necessary. Hyperspectral imaging (HSI) techniques are widely applied in plant phenotyping effectively measure traits caused by biotic or abiotic stresses. While previous HSI...
Accurate assessment of corn nitrogen content level is beneficial for growers to make informed fertilizing decisions save costs and optimize yields. Hyperspectral imaging technology has been shown be capable profiling plant's physiological statuses in a rapid noninvasive way. However, the spatial variance hyperspectral signal across leaf rarely analyzed. In this study, leveraging feature extraction ability deep learning models, both spectral information collected with handheld imager,...
By measuring correlated variables in housing estates corresponding to different grades of roads Zhengzhou City, obtain the conclusion based on correlation values: saturated headway is low, average travel speed high high-grade estate, i.e. impact types estate openness road passage. Then mode and scale, propose reasonable suggestions for urban planning sector traffic administrative department from two aspects: income increase expenditure reduction.