@article{zhan_zhou_bai_ge_2024, title={Bagging Improves the Performance of Deep Learning-Based Semantic Segmentation with Limited Labeled Images: A Case Study of Crop Segmentation for High-Throughput Plant Phenotyping}, volume={24}, ISSN={1424-8220}, url={http://dx.doi.org/10.3390/s24113420}, DOI={10.3390/s24113420}, abstractNote={Advancements in imaging, computer vision, and automation have revolutionized various fields, including field-based high-throughput plant phenotyping (FHTPP). This integration allows for the rapid and accurate measurement of plant traits. Deep Convolutional Neural Networks (DCNNs) have emerged as a powerful tool in FHTPP, particularly in crop segmentation-identifying crops from the background-crucial for trait analysis. However, the effectiveness of DCNNs often hinges on the availability of large, labeled datasets, which poses a challenge due to the high cost of labeling. In this study, a deep learning with bagging approach is introduced to enhance crop segmentation using high-resolution RGB images, tested on the NU-Spidercam dataset from maize plots. The proposed method outperforms traditional machine learning and deep learning models in prediction accuracy and speed. Remarkably, it achieves up to 40% higher Intersection-over-Union (IoU) than the threshold method and 11% over conventional machine learning, with significantly faster prediction times and manageable training duration. Crucially, it demonstrates that even small labeled datasets can yield high accuracy in semantic segmentation. This approach not only proves effective for FHTPP but also suggests potential for broader application in remote sensing, offering a scalable solution to semantic segmentation challenges. This paper is accompanied by publicly available source code.}, number={11}, journal={Sensors}, publisher={MDPI AG}, author={Zhan, Yinglun and Zhou, Yuzhen and Bai, Geng and Ge, Yufeng}, year={2024}, month={May}, pages={3420} }
 @article{bai_koehler-cole_scoby_thapa_basche_ge_2024, title={Enhancing estimation of cover crop biomass using field-based high-throughput phenotyping and machine learning models}, volume={14}, ISSN={1664-462X}, url={http://dx.doi.org/10.3389/fpls.2023.1277672}, DOI={10.3389/fpls.2023.1277672}, abstractNote={Incorporating cover crops into cropping systems offers numerous potential benefits, including the reduction of soil erosion, suppression of weeds, decreased nitrogen requirements for subsequent crops, and increased carbon sequestration. The aboveground biomass (AGB) of cover crops strongly influences their performance in delivering these benefits. Despite the significance of AGB, a comprehensive field-based high-throughput phenotyping study to quantify AGB of multiple cover crops in the U.S. Midwest has not been found. This study presents a two-year field experiment carried out in Eastern Nebraska, USA, to estimate AGB of five different cover crop species [canola ( Brassica napus L.), rye ( Secale cereale L.), triticale ( Triticale × Triticosecale L.), vetch ( Vicia sativa L.), and wheat ( Triticum aestivum L.)] using high-throughput phenotyping and Machine Learning (ML) models. Destructive AGB sampling was performed three times during each spring season in 2022 and 2023. An array of morphological, spectral, thermal, and environmental features from the sensors were utilized as feature inputs of ML models. Moderately strong linear correlations between AGB and the selected features were observed. Four ML models, namely Random Forests Regression (RFR), Support Vector Regression (SVR), Partial Least Squares Regression (PLSR), and Artificial Neural Network (ANN), were investigated. Among the four models, PLSR achieved the highest Coefficient of Determination (R 2 ) of 0.84 and the lowest Root Mean Squared Error (RMSE) of 892 kg/ha (Normalized RMSE (NRMSE) = 8.87%), indicating that PLSR could be the most appropriate method for estimating AGB of multiple cover crop species. Feature importance analysis ranked spectral features like Normalized Difference Red Edge (NDRE), Solar-induced Fluorescence (SIF), Spectral Reflectance at 485 nm (R485), and Normalized Difference Vegetation Index (NDVI) as top model features using PLSR. When utilizing fewer feature inputs, ANN exhibited better prediction performance compared to other models. Using morphological and spectral parameters as input features alone led to a R 2 of 0.80 and 0.77 for AGB prediction using ANN, respectively. This study demonstrated the feasibility of high-throughput phenotyping and ML techniques for accurately estimating AGB of multiple cover crop species. Further enhancement of model performance could be achieved through additional destructive sampling conducted across multiple locations and years.}, journal={Frontiers in Plant Science}, publisher={Frontiers Media SA}, author={Bai, Geng and Koehler-Cole, Katja and Scoby, David and Thapa, Vesh R. and Basche, Andrea and Ge, Yufeng}, year={2024}, month={Jan} }
 @article{bai_barker_scoby_irmak_luck_neale_schnable_awada_kustas_ge_2024, title={High-throughput physiological phenotyping of crop evapotranspiration at the plot scale}, volume={316}, ISSN={["1872-6852"]}, DOI={10.1016/j.fcr.2024.109507}, abstractNote={Platforms and instrumentation for Field High-Throughput Plant Phenotyping (FHTPP) are well developed to measure important traits for crop breeding and agronomic studies. However, the research has focused on morphological and spectral traits; and approaches to estimate major physiological processes such as evapotranspiration (ET) for small experimental plots are lacking. In this study, we put forward a new analytical framework to estimate plot-scale ET by integrating frequent phenotyping data (multispectral and thermal infrared images, canopy reflectance, and LiDAR point clouds) from a FHTPP system (known as NU-Spidercam), the weather data, a simplified two-source energy balance model, and reference ET and crop coefficient calculation. The new plot-scale ET method was tested on five field experiments involving maize and soybean crops over two growing seasons, with the different treatment levels of irrigation water. Estimated plot-scale ET was accumulated across the growing reason for each plot, and its association with grain yield was investigated with regression analysis. The result showed that plot-scale accumulated ET captured the seasonal trend of plot water use and clearly differentiated the irrigation treatments. Strong linear correlations were observed between plot-scale ET and grain yield, with R2 values ranging from 0.35 to 0.93 (average R2 = 0.71). Plot-scale ET appeared to be a more steady and stronger predictor of grain yield across the seasons than several other morphological and spectral traits including crop height, green pixel fraction, canopy temperature depression, and red-edge normalized difference vegetation index. High spatial and temporal resolution of the field phenotyping data, along with the new analytical framework reported, successfully estimated ET at small plot scale, which is difficult to achieve with other systems or methods. Our work of estimating ET at the plot-scale can be adopt to other ground-based platforms and drones, thus empowers physiologists, breeders, and agronomists for high-throughput phenotyping of water-use related traits and drought response evaluation.}, journal={FIELD CROPS RESEARCH}, author={Bai, Geng and Barker, Burdette and Scoby, David and Irmak, Suat and Luck, Joe D. and Neale, Christopher M. U. and Schnable, James C. and Awada, Tala and Kustas, William P. and Ge, Yufeng}, year={2024}, month={Aug} }
 @article{turc_sahay_haupt_de oliveira santos_bai_glowacka_2024, title={Up-regulation of non-photochemical quenching improves water use efficiency and reduces whole-plant water consumption under drought in <i>Nicotiana tabacum}, volume={75}, ISSN={0022-0957 1460-2431}, url={http://dx.doi.org/10.1093/jxb/erae113}, DOI={10.1093/jxb/erae113}, abstractNote={Abstract Water supply limitations will likely impose increasing restrictions on future crop production, underlining a need for crops that use less water per mass of yield. Water use efficiency (WUE) therefore becomes a key consideration in developing resilient and productive crops. In this study, we hypothesized that it is possible to improve WUE under drought conditions via modulation of chloroplast signals for stomatal opening by up-regulation of non-photochemical quenching (NPQ). Nicotiana tabacum plants with strong overexpression of the PsbS gene encoding PHOTOSYSTEM II SUBUNIT S, a key protein in NPQ, were grown under differing levels of drought. The PsbS-overexpressing lines lost 11% less water per unit CO2 fixed under drought and this did not have a significant effect on plant size. Depending on growth conditions, the PsbS-overexpressing lines consumed from 4–30% less water at the whole-plant level than the corresponding wild type. Leaf water and chlorophyll contents showed a positive relation with the level of NPQ. This study therefore provides proof of concept that up-regulation of NPQ can increase WUE, and as such is an important step towards future engineering of crops with improved performance under drought.}, number={13}, journal={Journal of Experimental Botany}, publisher={Oxford University Press (OUP)}, author={Turc, Benjamin and Sahay, Seema and Haupt, Jared and de Oliveira Santos, Talles and Bai, Geng and Glowacka, Katarzyna}, editor={Lawson, TracyEditor}, year={2024}, month={Mar}, pages={3959–3972} }
 @article{chamara_bai_ge_2023, title={AICropCAM: Deploying classification, segmentation, detection, and counting deep-learning models for crop monitoring on the edge}, volume={215}, journal={Computers and Electronics in Agriculture}, publisher={Elsevier}, author={Chamara, Nipuna and Bai, Geng and Ge, Yufeng}, year={2023}, pages={108420} }
 @article{bai_ge_2023, title={Crop Stress Sensing and Plant Phenotyping Systems: A Review}, volume={5}, DOI={10.12133/j.smartag.SA202211001}, number={1}, journal={Smart Agriculture}, author={Bai, G. and Ge, Y.}, year={2023}, pages={66–81} }
 @article{zhang_bai_chamara_ge_2023, title={Diurnal Variation of Canopy NDVI in Maize and Soybean}, journal={Authorea Preprints}, publisher={Authorea}, author={Zhang, Junxiao and Bai, Geng Frank and Chamara, Nipuna and Ge, Yufeng}, year={2023} }
 @inproceedings{zhang_thapa_chamara_bai_ge_2023, title={Estimating crop stomatal conductance from RGB, NIR, and thermal infrared images}, volume={12539}, booktitle={Autonomous Air and Ground Sensing Systems for Agricultural Optimization and Phenotyping VIII}, author={Zhang, Junxiao and Thapa, Kantilata and Chamara, Nipuna and Bai, Geng and Ge, Yufeng}, year={2023}, pages={107–114} }
 @article{bai_ge_leavitt_gamon_scoby_2023, title={Goniometer in the air: Enabling BRDF measurement of crop canopies using a cable-suspended plant phenotyping platform}, volume={230}, ISSN={1537-5110}, url={http://dx.doi.org/10.1016/j.biosystemseng.2023.04.017}, DOI={10.1016/j.biosystemseng.2023.04.017}, abstractNote={The Bidirectional Reflectance Distribution Function (BRDF) quantifies the distribution of the spectral reflectance of a target surface at various viewing and illumination angles. In-field measurement of the BRDF of vegetation canopies improves the characterization of off-nadir measurements and informs radiative transfer models of canopy reflectance, where the Lambertian assumption does not hold. However, current field goniometers are unable to measure BRDF efficiently, especially for tall vegetation across the growing season because of the limitations of clearance, field accessibility, and flexibility of the sensor field of view. In this study, we explored the potential of using a large-scale cable-suspended field phenotyping system to quantify BRDF of the canopy reflectance at selected bands and Vegetation Indices (VIs) of maize and soybean canopies. The system performance benefited from the following conditions: no crop damage, full-season field accessibility of tall canopies, automatic measurement, and accurate positioning. Correlation analysis shows that a strong correlation exists among the reflectance, VIs, and the Green Pixel Fraction. The hemispheric distributions of the spectral reflectance at selected bands and VIs were quantified at multiple dates. Hot spots were observed in the backscatter direction at visible and near-infrared (NIR) bands at the largest sensor zenith angle around the Solar Principal Plane (SPP). In contrast, cold spots of the Normalized Difference Vegetation Index (NDVI) and its related VIs were observed in the backscatter direction around solar zenith angles. The row effect was found for the maize canopy at the NIR band and the Near-infrared reflectance of vegetation (NIRv). NDVI had the lowest anisotropy index values among investigated bands and VIs. This system could be further leveraged to generate rapid and detailed BRDF data at a high spatiotemporal resolution using multiple sensors.}, journal={Biosystems Engineering}, publisher={Elsevier BV}, author={Bai, Geng and Ge, Yufeng and Leavitt, Bryan and Gamon, John A. and Scoby, David}, year={2023}, month={Jun}, pages={344–360} }
 @article{bhatti_heeren_o’shaughnessy_neale_larue_melvin_wilkening_bai_2023, title={Toward automated irrigation management with integrated crop water stress index and spatial soil water balance}, volume={24}, number={6}, journal={Precision Agriculture}, publisher={Springer US New York}, author={Bhatti, Sandeep and Heeren, Derek M and O’Shaughnessy, Susan A and Neale, Christopher MU and LaRue, Jacob and Melvin, Steve and Wilkening, Eric and Bai, Geng}, year={2023}, pages={2223–2247} }
 @article{chamara_islam_bai_shi_ge_2022, title={Ag-IoT for crop and environment monitoring: Past, present, and future}, volume={203}, journal={Agricultural systems}, publisher={Elsevier}, author={Chamara, Nipuna and Islam, Md Didarul and Bai, Geng Frank and Shi, Yeyin and Ge, Yufeng}, year={2022}, pages={103497} }
 @article{thapa_zhang_bai_ge_2022, title={Characterization of maize responses to differential nitrogen rates using image-based phenotyping}, journal={Authorea Preprints}, publisher={Authorea}, author={Thapa, Kantilata and Zhang, Junxiao and Bai, Geng Frank and Ge, Yufeng}, year={2022} }
 @article{zhang_chamara_thapa_bai_ge_2022, title={Estimating Maize and Soybean Stomatal Conductance Based on Time Series Soil Moisture, Canopy Temperature and Weather Conditions}, journal={Authorea Preprints}, publisher={Authorea}, author={Zhang, Junxiao and Chamara, Nipuna and Thapa, Kantilata and Bai, Geng and Ge, Yufeng}, year={2022} }
 @inproceedings{bai_scoby_suyker_neale_maguire_barker_ge_2022, title={Estimating evapotranspiration for crop breeding at the plot scale using energy balance and crop coefficients}, volume={2022}, booktitle={AGU Fall Meeting Abstracts}, author={Bai, Geng and Scoby, David and Suyker, Andy and Neale, Christopher MU and Maguire, Mitchell and Barker, Burdette and Ge, Yufeng}, year={2022}, pages={H25O–1285} }
 @inproceedings{nie_lunar_bai_ge_pitla_koksal_vuran_2022, title={mmWave on a Farm: Channel Modeling for Wireless Agricultural Networks at Broadband Millimeter-Wave Frequency}, booktitle={2022 19th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)}, author={Nie, Shuai and Lunar, Mohammad Mosiur and Bai, Geng and Ge, Yufeng and Pitla, Santosh and Koksal, Can Emre and Vuran, Mehmet C}, year={2022}, pages={388–396} }
 @inproceedings{ahm_alkady_jin_bai_samal_ge_2021, title={A deep convolutional neural network based image processing framework for monitoring the growth of soybean crops}, booktitle={2021 ASABE Annual International Virtual Meeting}, author={AHM, Nipuna Chamara and Alkady, Khalid H and Jin, Hongyu and Bai, Frank and Samal, Ashok and Ge, Yufeng}, year={2021}, pages={1} }
 @inbook{bai_ge_2021, title={Cable Suspended Large-Scale Field Phenotyping Facility for High-Throughput Phenotyping Research}, booktitle={High-Throughput Crop Phenotyping}, publisher={Springer International Publishing Cham}, author={Bai, Geng and Ge, Yufeng}, year={2021}, pages={39–53} }
 @article{bai_ge_2021, title={Crop Sensing and Its Application in Precision Agriculture and Crop Phenotyping}, journal={Fundamentals of Agricultural and Field Robotics}, publisher={Springer International Publishing}, author={Bai, Geng and Ge, Yufeng}, year={2021}, pages={137–155} }
 @article{singh_ge_heeren_walter-shea_neale_irmak_woldt_bai_bhatti_maguire_2021, title={Inter-relationships between water depletion and temperature differential in row crop canopies in a sub-humid climate}, volume={256}, journal={Agricultural Water Management}, publisher={Elsevier}, author={Singh, Jasreman and Ge, Yufeng and Heeren, Derek M and Walter-Shea, Elizabeth and Neale, Christopher MU and Irmak, Suat and Woldt, Wayne E and Bai, Geng and Bhatti, Sandeep and Maguire, Mitchell S}, year={2021}, pages={107061} }
 @inproceedings{wang_li_zhao_zhao_bai_ge_shi_2021, title={Investigate the potential of UAS-based thermal infrared imagery for maize leaf area index estimation}, volume={11747}, booktitle={Autonomous Air and Ground Sensing Systems for Agricultural Optimization and Phenotyping VI}, author={Wang, Lin and Li, Jiating and Zhao, Lin and Zhao, Biquan and Bai, Geng and Ge, Yufeng and Shi, Yeyin}, year={2021}, pages={1174703} }
 @inproceedings{singh_heeren_ge_bai_neale_maguire_bhatti_2021, title={Sensor-based irrigation of maize and soybean in East-Central Nebraska under a sub-humid climate}, booktitle={2021 ASABE Annual International Virtual Meeting}, author={Singh, Jasreman and Heeren, Derek M and Ge, Yufeng and Bai, Geng and Neale, Christopher MU and Maguire, Mitchell S and Bhatti, Sandeep}, year={2021}, pages={1} }
 @inproceedings{zhao_wang_li_bai_shi_ge_2021, title={Toward accurate estimating of crop leaf stomatal conductance combining thermal IR imaging, weather variables, and machine learning}, volume={11747}, booktitle={Autonomous Air and Ground Sensing Systems for Agricultural Optimization and Phenotyping VI}, author={Zhao, Lin and Wang, Lin and Li, Jiating and Bai, Geng and Shi, Yeyin and Ge, Yufeng}, year={2021}, pages={98–105} }
 @article{singh_heeren_ge_bai_2020, title={Capturing Spatial Variability in Maize and Soybean using Stationary Sensor Nodes}, author={Singh, Jasreman and Heeren, Derek M and Ge, Yufeng and Bai, Geng}, year={2020} }
 @article{lo_rudnick_dejonge_bai_nakabuye_katimbo_ge_franz_qiao_heeren_2020, title={Differences in soil water changes and canopy temperature under varying water × nitrogen sufficiency for maize}, volume={38}, ISSN={0342-7188 1432-1319}, url={http://dx.doi.org/10.1007/s00271-020-00683-2}, DOI={10.1007/s00271-020-00683-2}, abstractNote={Crop nitrogen (N) status is known to affect crop water status and crop water use. To investigate further the N effects on soil water changes and on canopy temperature, three water levels × four N levels were imposed on two growing seasons of maize in west central Nebraska, USA. Soil water changes were measured using a neutron probe, whereas canopy temperature was measured using infrared thermometers on a ground-based mobile platform. At all water levels, soil water losses over month-long intervals were generally greater as N levels increased. Given equal water levels, early afternoon canopy temperatures were usually lower with higher N levels, but no trend or even the opposite trend was occasionally observed. Jointly considering canopy reflectance and soil water depletion shows potential to explain much of the variation in estimated instantaneous water use among plots. However, determining the relative contributions of the canopy and soil factors on a particular day may require season-to-date knowledge of the crop. Further research on assimilating such sensor data for a combined stress coefficient would improve crop modeling and irrigation scheduling when variable water sufficiency and variable N sufficiency are simultaneously significant.}, number={5-6}, journal={Irrigation Science}, publisher={Springer Science and Business Media LLC}, author={Lo, Tsz Him and Rudnick, Daran R. and DeJonge, Kendall C. and Bai, Geng and Nakabuye, Hope Njuki and Katimbo, Abia and Ge, Yufeng and Franz, Trenton E. and Qiao, Xin and Heeren, Derek M.}, year={2020}, month={Aug}, pages={519–534} }
 @article{singh_heeren_rudnick_woldt_bai_ge_luck_2020, title={Soil structure and texture effects on the precision of soil water content measurements with a capacitance-based electromagnetic sensor}, volume={63}, number={1}, journal={Transactions of the ASABE}, publisher={American Society of Agricultural and Biological Engineers}, author={Singh, Jasreman and Heeren, Derek M and Rudnick, Daran R and Woldt, Wayne E and Bai, Geng and Ge, Yufeng and Luck, Joe D}, year={2020}, pages={141–152} }
 @article{lo_pringle_rudnick_bai_krutz_gholson_qiao_others_2020, title={Within-Field Variability in Granular Matrix Sensor Data and its Implications for Irrigation Scheduling}, volume={36}, ISSN={1943-7838}, url={http://dx.doi.org/10.13031/aea.13918}, DOI={10.13031/aea.13918}, abstractNote={Highlights Within-field variability was larger for individual depths than for the profile average across multiple depths. Distributions of the profile average were approximately normal, with increasing variances as the soil was drying. Probability theory was applied to quantify the effect of sensor set number on irrigation scheduling. The benefit of additional sensors sets may decrease for longer irrigation cycles and for more heterogeneous fields. Abstract. Even when located within the same field, multiple units of the same soil moisture sensor rarely report identical values. Such within-field variability in soil moisture sensor data is caused by natural and manmade spatial heterogeneity and by inconsistencies in sensor construction and installation. To better describe this variability, daily soil water tension values from 14 to 23 sets of granular matrix sensors during the middle part of four soybean site-years in the Mississippi Delta were analyzed. The soil water tension data were found to follow approximately normal distributions, to exhibit moderately high temporal rank stability, and to show strong positive correlation between mean and variance. Based on these observations and the existing literature, a probabilistic conceptual framework was proposed for interpreting within-field variability in granular matrix sensor data. This framework was then applied to investigate the impact of sensor set number (i.e., number of replicates) and irrigation triggering threshold on the scheduling of single-day and multi-day irrigation cycles. If a producer’s primary goal of irrigation scheduling is to keep soil water adequate in a particular fraction of land on average, the potential benefit from increasing sensor set number may be smaller than traditionally expected. Improvement, expansion, and validation of this probabilistic framework are welcomed for developing a practical and robust approach to selecting the sensor set number and the irrigation triggering threshold for diverse soil moisture sensor types in diverse contexts. Keywords: Irrigation scheduling, Probability, Sensors, Soil moisture, Soil water tension, Variability, Watermark.}, number={4}, journal={Applied Engineering in Agriculture}, publisher={American Society of Agricultural and Biological Engineers (ASABE)}, author={Lo, Tsz Him and Pringle, HC and Rudnick, Daran R and Bai, Geng and Krutz, L Jason and Gholson, Drew M and Qiao, Xin and others}, year={2020}, pages={437–449} }
 @article{yuan_wijewardane_jenkins_bai_ge_graef_2019, title={Early Prediction of Soybean Traits through Color and Texture Features of Canopy RGB Imagery}, volume={9}, ISSN={2045-2322}, url={http://dx.doi.org/10.1038/s41598-019-50480-x}, DOI={10.1038/s41598-019-50480-x}, abstractNote={Abstract Global crop production is facing the challenge of a high projected demand, while the yields of major crops are not increasing at sufficient speeds. Crop breeding is an important way to boost crop productivity, however its improvement rate is partially hindered by the long crop generation cycles. If end-season crop traits such as yield can be predicted through early-season phenotypic measurements, crop selection can potentially be made before a full crop generation cycle finishes. This study explored the possibility of predicting soybean end-season traits through the color and texture features of early-season canopy images. Six thousand three hundred and eighty-three images were captured at V4/V5 growth stage over 6039 soybean plots growing at four locations. One hundred and forty color features and 315 gray-level co-occurrence matrix-based texture features were derived from each image. Another two variables were also introduced to account for location and timing differences between the images. Five regression and five classification techniques were explored. Best results were obtained using all 457 predictor variables, with Cubist as the regression technique and Random Forests as the classification technique. Yield (RMSE = 9.82, R 2 = 0.68), Maturity (RMSE = 3.70, R 2 = 0.76) and Seed Size (RMSE = 1.63, R 2 = 0.53) were identified as potential soybean traits that might be early predictable.}, number={1}, journal={Scientific Reports}, publisher={Springer Science and Business Media LLC}, author={Yuan, Wenan and Wijewardane, Nuwan Kumara and Jenkins, Shawn and Bai, Geng and Ge, Yufeng and Graef, George L.}, year={2019}, month={Oct}, pages={14089} }
 @article{bai_ge_scoby_leavitt_stoerger_kirchgessner_irmak_graef_schnable_awada_2019, title={NU-Spidercam: A large-scale, cable-driven, integrated sensing and robotic system for advanced phenotyping, remote sensing, and agronomic research}, volume={160}, journal={Computers and Electronics in Agriculture}, publisher={Elsevier}, author={Bai, Geng and Ge, Yufeng and Scoby, David and Leavitt, Bryan and Stoerger, Vincent and Kirchgessner, Norbert and Irmak, Suat and Graef, George and Schnable, James and Awada, Tala}, year={2019}, pages={71–81} }
 @inproceedings{bai_ge_leavitt_gamon_qi_awada_graef_irmak_schnable_scoby_et al._2018, title={Capturing diurnal variation of phenotypic traits for breeding plots using NU-Spidercam}, booktitle={AGU Fall Meeting 2018, Dec 10-14, 2018, Washington D.C., USA.}, author={Bai, G and Ge, Y and Leavitt, B and Gamon, J and Qi, Y and Awada, T and Graef, G and Irmak, S and Schnable, J and Scoby, D and et al.}, year={2018} }
 @article{bai_jenkins_yuan_graef_ge_2018, title={Field-Based Scoring of Soybean Iron Deficiency Chlorosis Using RGB Imaging and Statistical Learning}, volume={9}, ISSN={1664-462X}, url={http://dx.doi.org/10.3389/fpls.2018.01002}, DOI={10.3389/fpls.2018.01002}, abstractNote={Iron deficiency chlorosis (IDC) is an abiotic stress in soybean that can cause significant biomass and yield reduction. IDC is characterized by stunted growth and yellowing and interveinal chlorosis of early trifoliate leaves. Scoring IDC severity in the field is conventionally done by visual assessment. The goal of this study was to investigate the usefulness of RGB (Red Green Blue) images of soybean plots captured under the field condition for IDC scoring. Sixty-four soybean lines with four replicates were planted in six fields over two years. Visual scoring (referred to as Field Score, or FS) was conducted at V3 – V4 growth stage; and concurrently RGB images of the field plots were recorded with a high-throughput field phenotyping platform. A second set of IDC scores was done on the plot images (displayed on a computer screen) consistently by one person in the office (referred to as Office Score, or OS). Plot images were then processed to remove weeds and extract six color features, which were used to train computer-based IDC scoring models (referred to as Computer Score, or CS) using Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM). The results showed that, in the fields where severe IDC symptoms were present, FS and OS were strongly positively correlated with each other, and both of them were strongly negatively correlated with yield. CS could satisfactorily predict IDC scores when evaluated using FS and OS as the reference (overall classification accuracy > 81%). SVM models appeared to outperform LDA models; and the SVM model trained to predict IDC OS gave the highest prediction accuracy. It was anticipated that coupling RGB imaging from the high-throughput field phenotyping platform with real-time image processing and IDC CS models would lead to a more rapid, cost-effective, and objective scoring pipeline for soybean IDC field screening and breeding.}, journal={Frontiers in Plant Science}, publisher={Frontiers Media SA}, author={Bai, Geng and Jenkins, Shawn and Yuan, Wenan and Graef, George L. and Ge, Yufeng}, year={2018}, month={Jul} }
 @article{bai_blecha_ge_walia_phansak_2017, title={Characterizing Wheat Response to Water Limitation Using Multispectral and Thermal Imaging}, volume={60}, ISSN={2151-0040}, url={http://dx.doi.org/10.13031/trans.11967}, DOI={10.13031/trans.11967}, abstractNote={Abstract. Effective screening of plant genotypes for their tolerance to abiotic stress is a vital step for crop improvement. Various sensing technologies can be used for developing automated plant phenotyping systems as well as for better control of stress levels imposed on the plants. In this study, seven different wheat genotypes (G1 through G7) were grown under two different water regimes in a greenhouse. Soil moisture was monitored by soil water tension sensors distributed among the experimental plots. A sensor platform with two cameras (a multispectral camera and a thermal infrared camera) was developed to capture top-view images of the wheat plots (once a week) during the course of the experiment. Image processing algorithms were developed to extract wheat growth index (GI) and crop water stress index (CWSI). Ratios of GI and CWSI between the two water treatments were calculated as proxies to assess drought tolerance of the different genotypes. The results showed that GI was correlated with NDVI measured by a GreenSeeker sensor (R 2 = 0.65). Among the seven wheat genotypes studied, G5, G6, and G7 exhibited consistently higher GI ratios and lower CWSI ratios, an indication of their superiority in drought tolerance. It was also found that non-uniform environmental conditions in the greenhouse had quite a large influence on wheat growth, which made the characterization and differentiation of phenotypes among various genotypes more challenging. It is concluded that the multispectral and thermal infrared imaging system has potential for phenotypic screening of wheat genotypes for drought tolerance in a semi-controlled environment. Keywords: CWSI, Drought, Image processing, Multispectral imaging, Phenotyping, Thermal infrared imaging.}, number={5}, journal={Transactions of the ASABE}, publisher={American Society of Agricultural and Biological Engineers (ASABE)}, author={Bai, Geng and Blecha, Sarah and Ge, Yufeng and Walia, Harkamal and Phansak, Piyaporn}, year={2017}, pages={1457–1466} }
 @inproceedings{ge_bai_irmak_awada_stoerger_graef_scoby_schnable_2017, title={High throughput field plant phenotyping facility at University of Nebraska-Lincoln and the first year experience}, volume={2017}, booktitle={AGU Fall Meeting Abstracts}, author={Ge, Yufeng and Bai, Geng and Irmak, Suat and Awada, Tala and Stoerger, Vincent and Graef, George and Scoby, David and Schnable, James}, year={2017}, pages={B51A–1771} }
 @article{bai_ge_hussain_baenziger_graef_2016, title={A multi-sensor system for high throughput field phenotyping in soybean and wheat breeding}, volume={128}, ISSN={0168-1699}, url={http://dx.doi.org/10.1016/j.compag.2016.08.021}, DOI={10.1016/j.compag.2016.08.021}, abstractNote={Collecting plant phenotypic data with sufficient resolution (in both space and time) and accuracy represents a long standing challenge in plant science research, and has been a major limiting factor for the effective use of genomic data for crop improvement. This is particularly true in plant breeding where collecting large-scale field-based plant phenotypes can be very labor intensive and costly. In this paper we reported a multi-sensor system for high throughput phenotyping in plant breeding. The system comprised five sensor modules (ultrasonic distance sensors, thermal infrared radiometers, NDVI sensors, portable spectrometers, and RGB web cameras) to measure crop canopy traits from field plots. A GPS was used to geo-reference the sensor measurements. Two environmental sensors (a solar radiation sensor and air temperature/relative humidity sensor) were also integrated into the system to collect simultaneous environmental data. A LabVIEW program was developed to control and synchronize measurements from all sensor modules and stored sensor readings in the host computer. Canopy reflectance spectra (by portable spectrometers) were post processed to extract NDVI and red-edge NDVI spectral indices; and RGB images were post processed to extract canopy green pixel fraction (as a proxy for biomass). The sensor system was tested in a soybean and wheat field trial. The results showed strong correlations among the sensor-based plant traits at both early and late growing season. Significant correlations were also found between the sensor-based traits and final grain yield at the early season (Pearson's correlation coefficient r ranged from 0.41 to 0.55) and late season (r from 0.55 to 0.70), suggesting the potential use of the sensor system to assist in phenotypic selection for plant breeding. The sensor system performed satisfactorily and robustly in the field tests. It was concluded that the sensor system could be a powerful tool for plant breeders to collect field-based, high throughput plant phenotyping data.}, journal={Computers and Electronics in Agriculture}, publisher={Elsevier BV}, author={Bai, Geng and Ge, Yufeng and Hussain, Waseem and Baenziger, P. Stephen and Graef, George}, year={2016}, month={Oct}, pages={181–192} }
 @inproceedings{ge_pandey_bai_2016, title={Estimating fresh biomass of maize plants from their RGB images in greenhouse phenotyping}, volume={9866}, booktitle={Autonomous Air and Ground Sensing Systems for Agricultural Optimization and Phenotyping}, author={Ge, Yufeng and Pandey, Piyush and Bai, Geng}, year={2016}, pages={8–13} }
 @article{ge_bai_stoerger_schnable_2016, title={Temporal dynamics of maize plant growth, water use, and leaf water content using automated high throughput RGB and hyperspectral imaging}, volume={127}, ISSN={0168-1699}, url={http://dx.doi.org/10.1016/j.compag.2016.07.028}, DOI={10.1016/j.compag.2016.07.028}, abstractNote={Automated collection of large scale plant phenotype datasets using high throughput imaging systems has the potential to alleviate current bottlenecks in data-driven plant breeding and crop improvement. In this study, we demonstrate the characterization of temporal dynamics of plant growth and water use, and leaf water content of two maize genotypes under two different water treatments. RGB (Red Green Blue) images are processed to estimate projected plant area, which are correlated with destructively measured plant shoot fresh weight (FW), dry weight (DW) and leaf area. Estimated plant FW and DW, along with pot weights, are used to derive daily plant water consumption and water use efficiency (WUE) of the individual plants. Hyperspectral images of plants are processed to extract plant leaf reflectance and correlate with leaf water content (LWC). Strong correlations are found between projected plant area and all three destructively measured plant parameters (R2 > 0.95) at early growth stages. The correlations become weaker at later growth stages due to the large difference in plant structure between the two maize genotypes. Daily water consumption (or evapotranspiration) is largely determined by water treatment, whereas WUE (or biomass accumulation per unit of water used) is clearly determined by genotype, indicating a strong genetic control of WUE. LWC is successfully predicted with the hyperspectral images for both genotypes (R2 = 0.81 and 0.92). Hyperspectral imaging can be a very powerful tool to phenotype biochemical traits of the whole maize plants, complementing RGB for plant morphological trait analysis.}, journal={Computers and Electronics in Agriculture}, publisher={Elsevier BV}, author={Ge, Yufeng and Bai, Geng and Stoerger, Vincent and Schnable, James C.}, year={2016}, month={Sep}, pages={625–632} }
 @article{yufeng ge_vincent stoerger_2016, title={Temporal dynamics of maize plant growth, water use, and leaf water content using automated high throughput RGB and hyperspectral imaging}, volume={127}, journal={Computers and Electronics in Agriculture}, publisher={Elsevier}, author={Yufeng Ge, Geng Bai and Vincent Stoerger, James C. Schnable}, year={2016}, pages={625–632} }
 @article{bai_nakano_ohashi_mizukami_yan_kramchote_2016, title={The influence of design parameters on the initial spray characteristics of the high-pressure air inclusion nozzle}, volume={26}, number={4}, journal={Atomization and Sprays}, publisher={Begel House Inc.}, author={Bai, Geng and Nakano, Kazuhiro and Ohashi, Shintaroh and Mizukami, Tomomichi and Yan, Haijun and Kramchote, Somsak}, year={2016} }
 @article{li_bai_yan_2015, title={Development and validation of a modified model to simulate the sprinkler water distribution}, volume={111}, journal={Computers and Electronics in Agriculture}, publisher={Elsevier}, author={Li, Yongchong and Bai, Geng and Yan, Haijun}, year={2015}, pages={38–47} }
 @article{kramchote_nakano_kanlayanarat_ohashi_takizawa_bai_2014, title={Rapid determination of cabbage quality using visible and near-infrared spectroscopy}, volume={59}, number={2}, journal={LWT-Food Science and Technology}, publisher={Academic Press}, author={Kramchote, Somsak and Nakano, Kazuhiro and Kanlayanarat, Sirichai and Ohashi, Shintaroh and Takizawa, Kenichi and Bai, Geng}, year={2014}, pages={695–700} }
 @phdthesis{bai_2014, title={Spray Performance of the High Pressure Air Inclusion Nozzle Used in Japan: In the aspects of the relative spray drift and the influence of key design parameters on spray performance}, school={Niigata University}, author={Bai, Geng}, year={2014} }
 @article{bai_nakano_mizukami_miyahara_ohashi_kubota_takizawa_yan_2013, title={Characteristics and classification of Japanese nozzles based on relative spray drift potential}, volume={46}, journal={Crop protection}, publisher={Elsevier}, author={Bai, Geng and Nakano, Kazuhiro and Mizukami, Tomomichi and Miyahara, Sumihiko and Ohashi, Shintaroh and Kubota, Yosuke and Takizawa, Ken-ichi and Yan, Haijun}, year={2013}, pages={88–93} }
 @inproceedings{bai_nakano_mizukami_miyahara_ohashi_takizawa_yan_others_2012, title={Nozzle classification system in Japan based on the relative spray drift potential.}, booktitle={Power and Machinery. International Conference of Agricultural Engineering-CIGR-AgEng 2012: agriculture and engineering for a healthier life, Valencia, Spain, 8-12 July 2012}, author={Bai, Geng and Nakano, K and Mizukami, T and Miyahara, S and Ohashi, S and Takizawa, K and Yan, HaiJun and others}, year={2012} }
 @article{bai_yan_2011, title={Effect of air drag coefficient on motion and evaporation of water droplet}, volume={42}, number={4}, journal={Shuili Xuebao(Journal of Hydraulic Engineering)}, publisher={Chinese Hydraulic Engineering Society,| a A-1 Fuxing Road| c Beijing| z …}, author={Bai, Geng and Yan, Hai-Jun}, year={2011}, pages={448–453} }
 @article{yan_bai_he_lin_2011, title={Influence of droplet kinetic energy flux density from fixed spray-plate sprinklers on soil infiltration, runoff and sediment yield}, volume={110}, number={2}, journal={Biosystems Engineering}, publisher={Academic Press}, author={Yan, HJ and Bai, G and He, JQ and Lin, G}, year={2011}, pages={213–221} }
 @article{bai_2011, title={Measuring Sprinkler Droplet Size with Modified Flour Methodology}, journal={Transactions of the Chinese Society for Agricultural Machinery}, author={Bai, Geng}, year={2011} }
 @article{yan_bai_he_li_2010, title={Model of droplet dynamics and evaporation for sprinkler irrigation}, volume={106}, number={4}, journal={Biosystems engineering}, publisher={Academic Press}, author={Yan, HJ and Bai, G and He, JQ and Li, YJ}, year={2010}, pages={440–447} }
 @article{yan_zhu_bai_yu_2009, title={Discussion on application of large-sized sprinkler irrigation machines in Inner Mongolia Autonomous region}, volume={1}, journal={Water Saving Irrigation}, author={Yan, H and Zhu, Y and Bai, G and Yu, P}, year={2009}, pages={18–21} }
 @article{bai_barker_scoby_irmak_luck_neale_schnable_heeren_awada_ge, title={Leveraging Proximal Imaging and Environmental Data for High-Resolution ET Modeling at Plot Scale}, journal={AGU23}, publisher={AGU}, author={Bai, Geng and Barker, Burdette and Scoby, David and Irmak, Suat and Luck, Joe D and Neale, Christopher MU and Schnable, James C and Heeren, Derek and Awada, Tala and Ge, Yufeng} }
 @article{bai, title={Spray Performance of the High Pressure Air Inclusion Nozzle Used in Japan}, author={Bai, Geng} }