The Ear Unwrapper: A Maize Ear Image Acquisition and Machine Learning Pipeline for Disease Severity Phenotyping

doi:10.21203/rs.3.rs-2728656/v1

Download PDF

Research Article

The Ear Unwrapper: A Maize Ear Image Acquisition and Machine Learning Pipeline for Disease Severity Phenotyping

https://doi.org/10.21203/rs.3.rs-2728656/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

To successfully breed any plant for disease resistance, an accurate method of phenotyping disease severity is crucial. Fusarium ear rot (FER) is a common disease of maize (Zea mays) caused by the pathogen Fusarium verticillioides (Sacc.) Nirenberg (synonym F. moniliforme Sheldon, teleomorph Gibberella moniliformis Wineland). Because of the quantitative nature of the disease, scoring disease severity is difficult and nuanced, relying on various ways to quantify damage caused by the pathogen. Towards the goal of designing a system with greater objectivity, reproducibility, and accuracy than subjective scores or estimations of area damaged, a system of semi-automated image acquisition and subsequent image analysis was designed.

Results

The tool created for image acquisition, “The Ear Unwrapper”, successfully obtains images of the full exterior of maize ears in roughly 10 seconds. To “unwrap” each ear of maize, the approach was to rotate the ear around its axis (the cob) and using a camera, take a continuous set of images of a single row of pixels and merge them to form one image. A set of images produced from The Ear Unwrapper were used to produce a probabilistic pixel classification model for predicting disease severity from unannotated images. The system was deliberately constructed using open-source software and off-the-shelf parts so that the image acquisition and analysis pipeline is adaptable for quantifying other maize ear pathogens, morphologies, and phenotypes. The data obtained from The Ear Unwrapper was correlated with two other phenotyping methods for validation and comparison and showed that the output from the system was reasonably accurate to determine lesion size.

Conclusions

This study provides an example of how a simplified image acquisition machine can be built and incorporated into a machine learning pipeline to measure phenotypes of interest. Here, The Ear Unwrapper was built to image ears of maize, but other cylindrical objects can also be “unwrapped” to obtain a single image of the object’s exterior. We also present how the use of machine learning in image analysis can be adapted from open-source software to estimate complex phenotypes, here, disease severity of Fusarium Ear Rot.

Semi-automated image acquisition

disease phenotyping

machine learning

Imaging of plant material for the purpose of phenotypic analysis continues to gain popularity in fields such as plant breeding, plant pathology, and plant systematics(1–4). On a larger scale, drone, airplane, and satellite imagery are used in large scale agriculture for monitoring a myriad of agronomic traits(5, 6). Imaging technology and the subsequent analysis allows for a reduction in labor and time-related costs while maintaining, if not improving, accuracy. The use of machine learning (ML) pipelines can decrease bias from human subjectivity, improve the efficiency of analysis pipelines, and increase the accuracy of predictions(7, 8). However, obtaining and preprocessing enough high-quality training data often creates a barrier to entry that is too high for labs or individuals to develop a robust, accurate, and efficient pipeline of image acquisition and analysis.

In plant pathology, accurate disease phenotyping results are used to guide day-to-day agricultural production (choice of plant variety, applications of pesticides, rogueing of plants), long-term breeding programs and molecular research(9–11). To acquire images for analysis, the type and scale of image is based on the purpose of the study. Those attempting to identify regions in a field under stress, image acquisition from drones, satellites, or other aerial tools are the most common(12). Conversely, studying individual plants requires close-up and high-resolution images, that are either compiled by hand or by using another device close to the target. Most studies using ML in plant pathology attempt to accurately identify (or classify) a microorganism, so images are often obtained with microscopes or hand-held tools(13, 14). Quantitative disease severity estimations that are required by plant breeding programs, are not limited to local image acquisition but instead depend on the scale of the study. Regardless, quantification of biotic or abiotic stressors is generally achieved by breaking down an image or set of images to smaller component parts (such as pixels) and using classification methods on those smaller parts before they are quantified(15, 16). This is often calculated or presented as a ratio to provide a consistent scale that is associated with plant health.

As it is a staple crop, maize (Zea mays), ML for image analysis has been used in a variety of studies where image acquisition methods are standardized(17–21). The disease addressed in this work is Fusarium ear rot (FER) caused by the pathogen Fusarium verticillioides, and the investigation focuses on the area of infected tissue on the ear of the maize plant. FER severity cannot be estimated without removing the husk of the ear and viewing the whole ear to observe the damage induced by the pathogen(22–24). Thus, traditional imaging of one side of a harvested ear is not sufficient to capture the information from the entire ear. In addition, high quality image acquisition of the ear is required to be able to distinguish between different types of damage or disease that may be present on the ear, whether for a ML pipeline or for traditional phenotyping.

The goal of the system presented here was to obtain an image of a full ear of maize in such a way that it could be used to quantify disease severity following inoculation. Our expectation would be that if we could “unwrap” the ear, such as one might remove the label from a can of food in order to convert a cylindrical object to a rectangular image, we would be able to accurately measure lesion size from all parts of the ear. As diseases (particularly FER) often occur in multiple places on a single ear or wraps around the ear, obtaining a full view of the ear was necessary for accurate phenotyping. The images produced by the machine presented here, named “The Ear Unwrapper”, were then analyzed with a ML pipeline estimating disease severity using probabilistic pixel classification. The Ear Unwrapper can be applied to any studies investigating maize ear phenotypes that require a full view of the ear.

Image acquisition and descriptive statistics:

Each maize ear, once mounted on the modified stepper motor, required 10 seconds for a full image to be taken. After the size correction and de-warping, images were roughly 1280 x 450–600 depending on the diameter of the ear. Over 3000 ears were imaged using The Ear Unwrapper which showed high reproducibility and no errors in image acquisition.

Correlation Of Severity Scores

Using the index of each photo, disease severity scores were adjusted and correlated in different ways. Log adjusted counts of white pixels (representing disease) from the manually annotated paint output were compared to the log adjusted pixel counts from the Ilastik prediction software generating an R² = 0.7409 (Fig. 1A). Next, log values of the pixel count for both Paint and Ilastik predictions were calculated then correlated with the log adjusted kernel counts. The log values of the Paint counts compared with the log adjusted kernel counts generated an R² = 0.4677 (Fig. 1B) and the log values of the Ilastik scores compared with the log kernel counts generated an R² = 0.2831 (Fig. 1C). Additional statistics from the output of all scoring methods can be found in Table 1.

Table 1

Statistics from the output of kernel counts, Ilastik predictions, and Paint annotations
Method	Average	Standard deviation	Minimum	Maximum
Log kernel count	1.48	0.22	0.85	1.90
Log Ilastik pixel count	3.78	0.67	0.78	4.62
Log Paint pixel count	3.87	0.39	2.81	4.62
Raw kernel count	33.9	16.9	7	80
Raw Ilastik pixel count	11622	10521	6	41290
Raw Paint pixel count	10270	7375	659	41290

The technology presented in this study demonstrates a method for quantifying disease symptoms by imaging maize ears via “unwrapping” the ear exterior to be viewed in a single rectangular image for different image analysis objectives. The Ear Unwrapper method of image stitching and de-warping of the different sized and shaped ears showed reliable, consistent, and accurate results in the downstream analysis. As phenotypic data acquisition is often the bottleneck that most breeding programs face, this study provides an example of how to remove this bottleneck by pairing commercially available hardware with machine learning. Here, we used images produced by The Ear Unwrapper in a machine learning pipeline to quantify the severity of damage caused by the Fusarium Ear Rot (FER) pathogen. As FER is an important disease of maize, accurate disease severity estimates are necessary to improve disease resistance in maize breeding programs. The Ear Unwrapper platform may also be used for a number of other diseases that infect maize ears, or for other investigations regarding the morphology and phenotypes of the maize ear.

Though the sample size of images chosen in this study was small (testing set N = 49), correlations between the results of different phenotypes were consistently positively correlated. While correlations between the log adjusted kernel counts and the log adjusted pixel counts were low (R² = 0.4677 and R² = 0.2831), this assumes a linear relationship between scoring methods. It is important to note that within the FER pathosystem, several methods of measuring disease severity exist, though none are universally used and accepted. However, when comparing the two methods that used the same images generated by the Ear Unwrapper, the correlation is much higher (R² = 0.7409), suggesting that the ML prediction pipeline is quite accurate for estimating disease area. By contrasting this high correlation with the lower correlation values between kernel counting and pixel counts, it could provide evidence that kernel counting does not correlate well with lesion area. But, again, neither method is accepted as superior for use in phenotyping disease severity, so additional work to determine the genetic basis of these different disease severity estimates should be done. The same size lesion on ears with different sized kernels would generate two different kernel counts but the same area (or number of pixels).

Regardless, the high and positive correlation between the images manually annotated in Paint and those generated from Ilastik demonstrate how image acquisition and analysis pipelines can accurately measure complex traits of interest such as FER infection severity using probabilistic pixel classification. As image segmentation for quantitative analysis is usually performed by complex deep learning neural networks, this evaluation of a quantitative trait by using a classification model provides an example of how more user-friendly software and other data-driven models can be used to obtain reasonable results(25–27). There are several options that may be used to improve this method. Increasing the number of images used for training will provide better resolution around the quality of the results, and possibly improve the calibration to provide more accurate predictions. Using more highly customized pixel classification methods or applying more computationally intensive neural network models could possibly provide better results, but instead, generating quantifiable data that can be objectively validated would give greater weight to images that increase correlations between predicted and actual scores. Further validations could be achieved using chemically validated data or pathosystems with well-established disease severity evaluation protocols that do not rely on expert knowledge or even visual estimations.

While The Ear Unwrapper can only image a single ear at a time, the same framework and image acquisition techniques could be applied using multiple ears/motors with only a slightly higher cost. This technology could also be adapted for applications using other wavelengths outside the visual spectrum, as the pipeline is mostly hardware-agnostic, provided the camera can accept external triggers for shutter control, and supports UVC device standards. Other imaging devices may be used if driver-level interfacing or hardware modification is provided. Similar technologies have been used for kernel analysis (counting and sorting), ear morphology (size, shape, and color), and molecular traits. The Ear Unwrapper, given the appropriate downstream image analysis pipeline, can be applied for these traits as well. While currently built to image maize ears, other cylindrical products that require analysis of the full image are easily adapted using the same framework. In this study, only ears with high (> 90%) seed set were included, but ears with low seed set and irregular morphologies were also imaged and produced high quality results (Supplemental Fig. 1).

The Ear Unwrapper provides a low-cost system that is built from readily available parts, which can be used to greatly improve the accuracy and efficiency of data acquisition, while also removing the considerable bias that is inherent in human analysis. For researchers and plant breeders seeking improved reproducibility of phenotyping assays, automating phenotyping with similar tools may be used to provide consistency of results and improve the accuracy of disease severity assessments.

FER disease inoculation:

A diversity panel of 66 sweetcorn and 2 field corn varieties of maize was grown in Citra, FL, from March-June of 2022. Ears were harvested approximately 21 days after 50% silking was observed in each variety. Ears were brought to an inoculation location before the husk was wiped down with 70% ethanol. Circles were drawn on the husk in the center of each ear to indicate the inoculation location. An 18-gauge needle was then used to make a hole through the husk to the cob. A Prima-tech® vaccinator was then primed with a spore solution and another 18-gauge needle was attached. Prior to inoculation, spore solutions were created by growing up isolates of Fusarium verticillioides in potato dextrose agar plates, taking five 1cm² mycelial plugs from the plates and placing them in ½ strength potato dextrose broth to shake for 5 days. After 5 days spore solutions were filtered through sterile cheese cloth and concentrated or diluted to equal 1x10⁵ conidia per ml. Once the spore solution was primed into the vaccinator, the needle was then placed into the existing hole in the ear and .75 ml of the spore solution was injected. Ears, once inoculated, were then placed in bags for humidity and isolation before being incubated at 27°C for 7 days. After 7 days, ears were removed from bags, husked, shanks were removed, and ears were individually placed on The Ear Unwrapper stepper motor. Often maize ears would have damage from other sources, primarily on the top of the ear. In these cases, the affected part of the ear was removed before being mounted on The Ear Unwrapper to minimize the confounding factors that would impact subsequent image analysis.

Hardware

To “unwrap” each ear of corn, the approach was to rotate the ear around its axis (the cob) and using a camera, take a continuous set of images to extract a single row of pixels from each to form one image. The motor used was a Nema 17 stepper motor that controls the position and rotation of the motor (and the object mounted on the motor) at a fraction of a degree. Here, each pulse of the stepper motor represents 0.45 degrees, thus 800 pulses are required to spin 360 degrees. The camera used was a See3CAM_24CUG camera (Fig. 2B). It was chosen because it has UVC driver compatibility that facilitates integration into an operating system as a USB webcam and because it has an external shutter trigger. The external trigger allows a microcontroller to control the shutter for synchronization with the stepper motor. As the motor is sent a pulse to rotate 0.45 degrees, the camera is also sent a pulse to capture an image. By syncing the pulses of the camera and the motor, one can control the speed while maintaining the accuracy of the image that is being produced via stitching. The pulses that are sent to the motor and the camera come from an Arduino UNO (SparkFun RedBoard) (Fig. 2C). Although reducing the time it took to obtain a single image was important, obtaining a clear, well lit, and focused image were all factors that reduced the frame rate. The USB bandwidth and stability of the corn ear when rotating also forced a reduction in frame rate.

The frame for the machine was built with Nylon and high-density polyethylene (HDPE) to provide durability for mounting the other components and allowing room for the ear to rotate less than one foot from the camera lens. The camera and motor were mounted on Picatinny rails. The stepper motor mount that came with the device was removed and replaced with two spikes (sharpened screws) that screwed into the existing stepper motor mount holes. Two spikes were used to provide the leverage required to rotate the ear on the sliding rails. A single conical spike of aluminum was fixed directly above the center of the modified stepper motor. This conical spike provided additional stability for the maize ear when mounted on the modified stepper spikes and allowed the ear to rotate freely (Fig. 2D).

To prepare ears to produce an image, ears were husked, the shanks cut or snapped off, silks removed, and the unpollinated tips of the ears removed. The Ear Unwrapper was placed in a photobox (Fotodiox™) to obtain images with consistent lighting and background (Fig. 2A). Ears were then mounted in the Ear Unwrapper by gently pressing the base of the ear onto the modified stepper motor spikes until it would stand without assistance and then sliding down the conical spike to the tip of the ear. The door to the photobox was closed and a custom Python script was used to capture, process, and output images.

Calibration And Error Correction

The 24CUG camera uses a wide-angle lens with focal length 3.0mm and aperture f/2.8, a benefit because the camera could be positioned closer to the maize ear while keeping the full ear in view. To correct for the “fisheye” effect that results, a correction algorithm was used to remove this distortion and extract the rectangular image. This was done in Python with the OpenCV library and calibrated by printing out a checkerboard and photographing it with the camera at the fixed distance from the camera to the motor. This also allowed for the determination of pixels/cm measurement. The de-warped image of the checkerboard was used to produce a correction matrix that was then applied to every image captured by the camera. To more efficiently use the sensor of the 24CUG camera default 1280 x 720 resolution, we rotated the camera 90 degrees so that the 1280 resolution was vertical and could better accommodate the geometry of the corn ears. When an image was taken, a single vertical line of 1280 pixels was obtained before the motor turns 0.45 degrees and another image is taken, and the next vertical line was stitched to the previous line of pixels. This was repeated 800 times until the ear completed a 360-degree rotation. This value was based on an estimation that an average ear of corn is roughly 800 pixels or 2 inches in diameter when viewed at the specified distance and resolution from the camera lens. Although there are substantial morphological differences in ears of corn that could cause the image to be stretched or compressed, particularly when using a diversity panel, this was addressed by capturing an additional four images, each 200 pulses (or 90 degrees) apart. These four additional images captured the full field of view rather than a line of pixels. Segmenting the ear from these four images allowed us to calculate an estimate of the ear diameter, and correct distortions in the 'unwrapped’ image incurred when imaging ears that were significantly wider or narrower than the 2” ear diameter the image dimensions were calibrated for.

Image Processing

A custom Python application was built to interface with the UVC driver, control the camera shutter and the stepper motor (via the Arduino), and capture the images. This application also provided a command-line interface to allow user control for photo capture and file naming. The photos were captured, named, and saved according to the date, time, and an ID number entered by the user as a unique identifier. Images were saved as PNG files and then filtered and sorted based on a set of criteria. First, images of ears with < 50% pollination, significant damage that was not caused by F. verticillioides, or images that were blurred were removed as these factors influence the ability to measure disease severity. After removing and filtering the images, 59 images remained before 10 images were selected to train the model. When assigning images to the training folder, the images were chosen based on morphological and phenotypic diversity to provide a range of disease severity, ear size, and a diversity of lesion colors and shapes. Images used for training were also selected so that visual differentiation between healthy and diseased tissues was clear. The testing set consisted of the remaining 49 images.

Pixel Classification

To build and train a classification model, we used the interactive machine learning software Ilastik(28). This software provides a pre-defined feature space across several workflows for biological image analysis, reducing both the amount of computing time and training data required to develop an accurate classifier. We used the 'pixel classification' workflow to semantically segment the images according to our three class labels - background, healthy tissue, and disease tissue. The background consisted of all image areas which were not ear tissue, and all maize ear tissues without fungal hyphae were considered healthy. Any tissue that contained visually apparent disease symptoms was annotated as diseased tissue (Supplemental Fig. 2). We used the default Random Forest classifier, and feature selection settings of σ3 (Sigma of 1.60) for Color/Intensity, Edge, and Texture. The ten training images were imported to Ilastik, and each image was manually annotated with the class labels until the real-time classifier predictions became stable (i.e. when additional annotation caused either no visible change or minimal change in the model feedback). Once training was complete, the remaining 49 testing/validation images were imported for batch processing. The model was used to generate a probability map for each image, i.e. an image with identical dimensions where each pixel is colored by the classifier according to the class label that it most likely represents in the original image (Fig. 3B).

Imagej And Opencv Processing

Outputs from the testing data set were processed in ImageJ (FIJI), and OpenCV using a custom Python script(29, 30). Probability files were converted from TIFF to PNG and a median blur function (Ksize = 11) was applied to each image in ImageJ. In OpenCV, the RGB images were converted to an HSV (hue, saturation, and value) representation and thresholded to create a mask for the pixels classified as likely to be disease. This mask image was converted to binary (black and white) where disease is represented as white pixels (Fig. 3D,E). These white pixels were then counted using a Python script and written out to a CSV file with their corresponding unique ID numbers for reference to their manually recorded scores.

Scoring Methods And Correlations

Ears that were imaged were also scored in two different ways. The first was a user with experience with FER infections counting the kernels that had visual symptoms consistent with FER. The second was by manually annotating each image in the program Microsoft Paint (on Windows 10) by coloring the disease pixels with the same blue color used in Ilastik, using the same methods to blur, threshold, and mask the image as was done to the Ilastik output images (Fig. 3C,E). It should also be noted that the expert counting diseased kernels was the same person manually annotating the images on Paint to provide continuity. To correlate the kernel counting scores with the predicted pixel counts from Ilastik and manually annotated pixel counts from Paint, the log scores of each were calculated and compared.

FER: Fusarium Ear Rot

ML: Machine Learning

HDPE: high-density polyethylene

Ethics approval: NA

Consent for publication: NA

Availability of data and materials:

The software used to obtain the data used in this study can be found at the Github link: https://github.com/ohudson1/The-Ear-Unwrapper-.git

The raw dataset used to generate these results can be found in Supplemental Table 1.

Competing interests: The authors declare that they have no competing interests.

Funding: This research was funded by the University of Florida and the United States Department of Agriculture–National Institute of Food and Agriculture Tactical Sciences for Agricultural Biosecurity Institute of Food and Agricultural Sciences Project 13117320

Author contributions:

JB and OH initiated the project. OH acquired the data. OH, CB, and DH contributed to the design for The Ear Unwrapper. CB and DH built hardware and accompanying software for The Ear Unwrapper. JB oversaw the project and secured all funding. All authors read and approved the final manuscript.

Acknowledgments:

Dr. Marcio Resende and the Sweetcorn lab for assistance with field trials

Das Choudhury S, Samal A, Awada T. Leveraging image analysis for high-throughput plant phenotyping. Front Plant Sci. 2019 Apr 16;10.
Hartmann A, Czauderna T, Hoffmann R, Stein N, Schreiber F. HTPheno: An image analysis pipeline for high-throughput plant phenotyping. BMC Bioinformatics. 2011 May 12;12.
Fida F, Talib M, Fida F, Iqbal N, … MAPB, 2018 undefined. Leaf image recognition based identification of plants: Supportive framework for plant systematics. PSM Biological Research [Internet]. 2018 [cited 2023 Feb 7];3(3):125–31. Available from: https://core.ac.uk/download/pdf/327166021.pdf
Gobalakrishnan N, … KP… signal processing, 2020 undefined. A systematic review on image processing and machine learning techniques for detecting plant diseases. International Conference on Communication and Signal Processing [Internet]. 2020 [cited 2023 Feb 7]; Available from: https://ieeexplore.ieee.org/abstract/document/9182046/
Zhang C, Marzougui A, Sankaran S. High-resolution satellite imagery applications in crop phenotyping: An overview. Comput Electron Agric. 2020 Aug 1;175:105584.
van der Merwe D, Burchfield DR, Witt TD, Price KP, Sharda A. Drones in agriculture. Advances in Agronomy. 2020 Jan 1;162:1–30.
Katarya R, Raturi A, Mehndiratta A, Thapper A. Impact of Machine Learning Techniques in Precision Agriculture. Proceedings of 3rd International Conference on Emerging Technologies in Computer Engineering: Machine Learning and Internet of Things, ICETCE 2020. 2020 Feb 1;18–23.
Sharma A, Jain A, Gupta P, Chowdary V. Machine Learning Applications for Precision Agriculture: A Comprehensive Review. IEEE Access. 2021;9:4843–73.
Mutka AM, Bart RS. Image-based phenotyping of plant disease symptoms. Front Plant Sci. 2015 Jan 5;5(JAN):734.
Shakoor N, Lee S, Mockler TC. High throughput phenotyping to accelerate crop breeding and monitoring of diseases in the field. Curr Opin Plant Biol. 2017 Aug 1;38:184–92.
Mahlein AK. Plant disease detection by imaging sensors – Parallels and specific demands for precision agriculture and plant phenotyping. Plant Dis [Internet]. 2016 Feb 1 [cited 2023 Feb 7];100(2):241–54. Available from: https://apsjournals.apsnet.org/doi/10.1094/PDIS-03-15-0340-FE
Gago J, Douthe C, Coopman RE, Gallego PP, Ribas-Carbo M, Flexas J, et al. UAVs challenge to assess water stress for sustainable agriculture. Agric Water Manag. 2015 May 1;153:9–19.
Cheshkova AF. A review of hyperspectral image analysis techniques for plant disease detection and identif ication. Vavilov Journal of Genetics and Breeding [Internet]. 2022 [cited 2023 Feb 7];26(2):202. Available from: /pmc/articles/PMC8983301/
Perez-Sanz F, Navarro PJ, Egea-Cortines M. Plant phenomics: an overview of image acquisition technologies and image data analysis algorithms. Gigascience [Internet]. 2017 Nov 1 [cited 2023 Feb 7];6(11):1–18. Available from: https://academic.oup.com/gigascience/article/6/11/gix092/4316962
Pantazi XE, Moshou D, Oberti R, West J, Mouazen AM, Bochtis D. Detection of biotic and abiotic stresses in crops by using hierarchical self organizing classifiers. Precis Agric [Internet]. 2017 Jun 1 [cited 2023 Feb 9];18(3):383–93. Available from: https://link.springer.com/article/10.1007/s11119-017-9507-8
Singh A, Ganapathysubramanian B, Singh AK, Sarkar S. Machine Learning for High-Throughput Stress Phenotyping in Plants. Trends Plant Sci. 2016 Feb 1;21(2):110–24.
Wiesner-Hanks T, Stewart EL, Kaczmar N, Dechant C, Wu H, Nelson RJ, et al. Image set for deep learning: Field images of maize annotated with disease symptoms. BMC Res Notes [Internet]. 2018 Jul 3 [cited 2023 Feb 9];11(1):1–3. Available from: https://bmcresnotes.biomedcentral.com/articles/10.1186/s13104-018-3548-6
Saju KK. Classification of Macronutrient Deficiencies in Maize Plant Using Machine Learning Keyword: Deep learning with auto encoders Feature extraction K nearest neighbors Nutritional deficiency Support vector machines. International Journal of Electrical and Computer Engineering (IJECE). 2018;8(6):4197–203.
Han L, Yang G, Dai H, Xu B, Yang H, Feng H, et al. Modeling maize above-ground biomass based on machine learning approaches using UAV remote-sensing data. Plant Methods [Internet]. 2019 Feb 4 [cited 2023 Feb 9];15(1):1–19. Available from: https://plantmethods.biomedcentral.com/articles/10.1186/s13007-019-0394-z
Tu K, Wen S, Cheng Y, Xu Y, Pan T, Hou H, et al. A model for genuineness detection in genetically and phenotypically similar maize variety seeds based on hyperspectral imaging and machine learning. Plant Methods [Internet]. 2022 Dec 1 [cited 2023 Feb 9];18(1):1–17. Available from: https://plantmethods.biomedcentral.com/articles/10.1186/s13007-022-00918-7
Panigrahi KP, Das H, Sahoo AK, Moharana SC. Maize Leaf Disease Detection and Classification Using Machine Learning Algorithms. Advances in Intelligent Systems and Computing [Internet]. 2020 [cited 2023 Feb 9];1119:659–69. Available from: https://link.springer.com/chapter/10.1007/978-981-15-2414-1_66
Munkvold GP, Hellmich RL, Showers WB. Reduced Fusarium Ear Rot and Symptomless Infection in Kernels of Maize Genetically Engineered for European Corn Borer Resistance. https://doi.org/101094/PHYTO199787101071 [Internet]. 2007 Feb 20 [cited 2023 Feb 9];87(10):1071–7. Available from: https://apsjournals.apsnet.org/doi/10.1094/PHYTO.1997.87.10.1071
Munkvold GP. Epidemiology of Fusarium diseases and their mycotoxins in maize ears. Eur J Plant Pathol [Internet]. 2003 Sep [cited 2023 Feb 9];109(7):705–13. Available from: https://link.springer.com/article/10.1023/A:1026078324268
Mesterházy Á, Lemmens M, Reid LM. Breeding for resistance to ear rots caused by Fusarium spp. in maize – a review. Plant Breeding [Internet]. 2012 Feb 1 [cited 2023 Feb 9];131(1):1–19. Available from: https://onlinelibrary.wiley.com/doi/full/10.1111/j.1439-0523.2011.01936.x
Midtvedt B, Helgadottir S, Argun A, Pineda J, Midtvedt D, Volpe G. Quantitative digital microscopy with deep learning. Appl Phys Rev [Internet]. 2021 Feb 19 [cited 2023 Feb 9];8(1):011310. Available from: https://aip.scitation.org/doi/abs/10.1063/5.0034891
Van Valen DA, Kudo T, Lane KM, Macklin DN, Quach NT, DeFelice MM, et al. Deep Learning Automates the Quantitative Analysis of Individual Cells in Live-Cell Imaging Experiments. PLoS Comput Biol [Internet]. 2016 Nov 1 [cited 2023 Feb 9];12(11):e1005177. Available from: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005177
Ebrahim SA, Poshtan J, Jamali SM, Ebrahim NA. Quantitative and Qualitative Analysis of Time-Series Classification Using Deep Learning. IEEE Access. 2020;8:90202–15.
Berg S, Kutra D, Kroeger T, Straehle CN, Kausler BX, Haubold C, et al. ilastik: interactive machine learning for (bio)image analysis. Nature Methods 2019 16:12 [Internet]. 2019 Sep 30 [cited 2023 Feb 7];16(12):1226–32. Available from: https://www.nature.com/articles/s41592-019-0582-9
Bradski G. The OpenCV Library. Dr Dobb’s Journal of Software Tools. 2000;
Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature Methods 2012 9:7 [Internet]. 2012 Jun 28 [cited 2023 Feb 9];9(7):676–82. Available from: https://www.nature.com/articles/nmeth.2019

No competing interests reported.

Supplementalfigure1.jpg
Supplemental Figure 1.Examples of morphologically diverse maize sample images produced by The Ear Unwrapper.
Supplementalfigure2.jpg
Supplemental Figure 2. Example of annotation in Ilastik. Yellow lines are drawn on the background. Blue lines are drawn on healthy tissue, and red lines are drawn on disease tissue.
SupplementaryTable1.csv
Supplemental Table 1. All raw data values that were used in generating results.

Download PDF

Version 1

posted

You are reading this latest preprint version

The Ear Unwrapper: A Maize Ear Image Acquisition and Machine Learning Pipeline for Disease Severity Phenotyping

Status:

Version 1

Abstract

Background

Results

Conclusions

Figures

Background

Results

Image acquisition and descriptive statistics:

Correlation Of Severity Scores

Discussion

Conclusions

Methods

FER disease inoculation:

Hardware

Calibration And Error Correction

Image Processing

Pixel Classification

Imagej And Opencv Processing

Scoring Methods And Correlations

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1