Browsing by Subject "phenotyping"
Now showing 1 - 8 of 8
Results Per Page
Sort Options
Item Development of Image Analysis Tools to Quantify Potato Tuber Quality Traits(2022-08) Miller, MichaelPotato is the most popular non-cereal food crop and a major staple crop. Despite the importance of potato, it has seen little yield improvement through breeding over the past century when compared to other crops. One difficulty in potato breeding is the large number of quality traits that must be accounted for in order to create marketable potato varieties. These quality traits are often measured using imprecise, subjective scales. This thesis covers my work in improving the tools available for use in measuring and breeding for potato tuber quality traits. In Chapter 1, I review the literature relevant to a selection of quality traits and their measurement. I discuss machine learning and its use in identifying more intricate tuber quality traits, as well as efforts to perform genomic selection in autotetraploid potato as a possible application for highly quantitative quality trait data. Chapter 2 covers the mechanics and capabilities of the potato tuber image analysis program, TubAR. I compare the quantitative measurements provided by TubAR to human visual scores for analogous traits. In Chapter 3, I discuss efforts to expand the scope of traits able to be measured with image analysis by employing machine learning image classification, using the pressure bruise and skin finish traits.Item High-resolution grape cluster images and color-based segmentations for population GE1025 in 2017 and 2018(2019-04-18) Underhill, Anna N; underhillanna@gmail.com; Underhill, Anna NThis is a collection of images collected in 2017 and 2018 from the UMN grape research population GE1025 (parentage described in Teh et al., 2017) that was used for two image analysis-based phenotyping experiments: one involving cluster compactness, and one involving skin color. Contained in each year are RAW images, color-corrected images, and segmented images for berry, stem, and background. Details about methods used to capture and segment images are available at https://github.com/underhillanna/GrapeImageAnalysis, or in Underhill, 2019.Item Hirsch Lab UAV Commercial Maize Phenotyping Project at UMN SROC Waseca: 2020, 2021, and 2022(2024-04-22) Sweet, Dorothy D; Hirsch, Candice N; Hirsch, Cory D; cnhirsch@umn.edu; Hirsch, Candice N; Candice Hirsch Lab; Cory Hirsch LabThis dataset provides a valuable resource for evaluating the ability of unoccupied aerial vehicles to collect plant height information from commercial agricultural fields and predict within field variation in yield using temporal traits including plant height, growth rate, and vegetative indices. Many flights were conducted over commercial maize fields using an UAV equipped with an RGB camera and this dataset includes orthomosaics and digital elevation models generated from those flights as well as plot boundary shape files used for extraction of data from those flights. Data in this repository includes extracted plant height, extracted RGB vegetative indices, manual height measurements, weather data, soil data, and grain yield. This experiment consisted of three commercial fields containing single maize hybrids and is therefore useful in assessing the ability of UAV extracted values in identifying within field variation for prediction of yield. It can also be used to test different methods of extracting plant height values from commercial fields as it includes manual measurements of height to be used in evaluation.Item Investigating the Capability of Hyperspectral Imaging for the Estimation of Wheat Leaf Rust Disease(2016-09-03) Crowell, Olivia LDetecting disease in crops increases yields and reduces economic loss. Traditional methods detect plant disease severity by hand, but it is a slow, time-consuming, and subjective process. Because there are physical, chemical, and physiological changes in plants with diseases, current research focuses on optical imaging as a more useful technique for monitoring disease. This experiment uses hyperspectral imaging (HSI) to investigate its capability as a diagnostic tool for wheat leaf rust disease. HSI has the potential to create more efficient High Throughput Phenotyping (HTP) methods to assist wheat breeders in the selection of a more resistant wheat variety. To collect data for this purpose, a HSI camera was attached to a ground vehicle that scanned wheat plots on an experimental field in St. Paul a month after the fungus was inoculated into the plots. White panels were laid out into the field to normalize the radiance based on the irradiance. Preprocessing techniques converted the pixels from their digital number to reflectance, which measures any physiological changes. The average spectral signatures of pixels were then compared to scoring data on a spectrum of R (resistant) to S (susceptible). Other categories include MR (moderately resistant), MS (moderately susceptible), etc. However, the spectral signatures that matched with R, MR, MRMS, MS, S, etc. varied unreasonably. R and S spectral lines were very similar to each other and did not exist as a maximum and minimum. This study is inconclusive which could be due to the calibration process since light cannot be controlled perfectly. It could also be because of the lack of control with varied wheat during data collection times.Item Sample 360 video for the analysis of plant movement(2018-09-12) Susko, Alexander, Q; susko004@umn.edu; Susko, Alexander, Q; University of Minnesota Oat Breeding and Genetics LabViolent movement of cereal crop stems can lead to failure under high winds. Known as lodging, this phenomenon is particularly severe in cereal crops such as oat, barley, and wheat, and contributes to yield and economic losses. Quantifying the movement of cereal crops under field wind stress could aid in the breeding and selecting of lodging resistant cereals. We present a method to quantify the wave like movement of cereal crop rows in a high throughput fashion under field wind conditions. By analyzing pre-defined regions of hemispherical 4K resolution video, we obtain a time varying color signal of wind induced stem and canopy movement. Bandpass filtering is applied to the color signals to filter out changes in lighting due to sunlight changes, enabling comparisons across different lighting conditions. Peaks are then identified in the signal, and the distance in frames to the next peak as well as the absolute area under the curve between peaks is recorded. The distributions of distances to adjacent peaks (expressed as frequencies) are recorded and the area within a defined frequency bin is summed to get an approximation of the frequency and amount movement. We applied this method to analyze the wind induced movement of 16 cereal cultivars planted in a randomized complete block design on 5 different windy days. We detected significant differences in the mean frequency and amplitude within 0.2 Hz frequency bins among 16 cereal cultivars, with mean frequencies ranging between 1.24 and 1.53 Hz. This method quantifies the frequency and amplitude of movement in cereal varieties at high throughput in the field, and shows promise for characterizing the physiological basis for differences in cereal movement and lodging resistance.Item Springer Lab UAV Maize Phenotyping Project at UMN StPaul: 2018 and 2019(2020-05-05) Tirado, Sara B; Hirsch, Candice N; Springer, Nathan M; springer@umn.edu; Springer, Nathan M; Springer LabThis dataset provides a valuable resource for evaluating the utility of unmanned aerial vehicles to collect phenotypic data in agricultural fields. Many flights throughout the growing season of a maize experiment were conducted and this dataset includes digital elevation models generated from images within these flights, the plot boundary shapefiles for plot identification, plant height values extracted following Tirado et al., 2019 procedure, hand measurement height values conducted following flights, and yield data for each plot. This maize experiment consisted of twelve hybrids planted at three different planting densities (low, medium and high) and two planting dates (early and late) across two years and therefore provides a valuable resource for evaluating how temporal data collected from UAVs can aid in assessing plant productivity. It can also be utilized to develop and test different protocols for plant height extraction from DEMs at different growth stages as the hand measurements can be used to test the accuracy.Item Temporally resolved growth patterns in diverse maize panel(2023-01-27) Sweet, Dorothy D; Tirado, Sara B; Cooper, Julian S; Springer, Nathan M; Hirsch, Cory D; Hirsch, Candice N; cnhirsch@umn.edu; Hirsch, Candice N; Candice Hirsch Lab; Cory Hirsch LabPlant height is used in many breeding programs for assessing plant health across environments and predicting yield, which can be used in identifying superior hybrids or evaluating abiotic stress factors. This has often been measured at a single time point when plants have reached their terminal height for the season. Collection of plant height using unoccupied aerial vehicles (UAVs) is faster, allowing for measurements throughout the growing season which could facilitate a better understanding of plant-environment interaction and responses. To assess variation in plant height and growth rate throughout development, plant height data was collected weekly for a panel of ~500 diverse inbred lines over four growing seasons. The variation in plant height throughout the season was found to be significantly explained by genotype, year, and genotype-by-year interactions throughout vegetative growth. However, the relative contributions of these different sources of variation fluctuated throughout development. This variation was further captured by Fréchet distance values which identified genotypes with consistently high or low distances in each of the four years - high distance genotypes being more dissimilar between replications and therefore capturing more environmental variation. Genome-wide association studies revealed many significant SNPs associated with plant height and growth rate at different parts of the growing season that would not be identified by terminal height alone. When comparing growth rates estimated from plant height to growth rates estimated from another morphological characteristic, canopy cover, we found greater stability in growth curves estimated by plant height. This potentially makes canopy cover more useful for understanding environmental modulation of overall plant growth and plant height better for understanding genotypic modulation of overall plant growth. Overall, this suggests evaluations of plant growth throughout the season provide more information than terminal plant height alone.Item USDA-ARS Phenocart RGB Imagery Collected in Brookings, SD in 2021(2024-06-13) Ewing, Patrick; Runck, Bryan; patrick.ewing@usda.gov; Ewing, Patrick; Real-time Geoinformation Systems Lab, GEMS Informatics CenterData were collected from an experimental field in 2021 at the Eastern South Dakota Soil and Water Research Farm in Brookings, SD, USA (44.351 N, 96.805 W). The experiment consisted of a number of oat (Avena sativa L.) variety-by-seeding-rate treatments that were further divided into medium red clover planting treatments in a strip-block design with four replicates and a plot size of 6 m by 6 m. Oat treatments crossed variety (Reins, Natty, Sumo) and target oat population (140, 220, and 320 plants m-2) in 19 cm, drilled rows; red clover showed no responses to these oat treatments. Red clover treatments compared clover planted concurrently with oats (“underseeded”) on April 28, 2021; planted after oat harvest (“post-harvest”) on August 12, 2021; or no clover (“fallow”). Red clover was drilled at 1.25 cm depth at a rate of 8.2 kg ha-1 at a row spacing of 19 cm. An herbicide application of 210 g ha-1 sethoxydim (Poast, BASF Crop Protection, Research Triangle Park, North Carolina, USA), which selectively targets monocots, was applied on August 20, 2021, to control volunteer oats. A total of 720 RGB JPEG images were collected over six dates. The dates span the emergence of the post-harvest red clover planting to the first killing frost: August 21st, September 9th, September 29th, October 5th, October 15th, and October 25th. Images were collected by a Canon PowerShot ELPH 190 IS at a height of 2.5 m in the center of each plot using a phenotyping cart (White & Conley, 2013). To mirror the simplest use by researchers and practitioners, the camera was configured in full default, automatic mode, including ISO (a standard setting for controlling image darkness) and white balance and a 74-degree horizontal field of view. One image was taken from a representative location per plot on each date.