Evaluation of cotton emergence using UAV-based imagery and deep learning

Abstract Crop emergence is an important agronomic factor for making field management decisions, such as replanting, that are time-sensitive and need to be made at very early stages. Crop emergence, evaluated using plant population, stand count and uniformity, is conventionally quantified manually, not accurate, and labor and time intensive. Unmanned aerial vehicle (UAV)-based imaging systems are able to scout crop fields rapidly. However, data processing can be too slow to make timely decision making. The goal of this study was to develop a novel image processing method for processing UAV images in nearly real-time. In this study, a UAV imaging system was used to capture RGB image frames of cotton seedlings to evaluate stand count and canopy size. Images were pre-processed to correct distortions, calculate ground sample distance and geo-reference cotton rows in the images. A pre-trained deep learning model, resnet 18, was used to estimate stand count and canopy size of cotton seedlings in each image frame. Results showed that the developed method could estimate stand count accurately with R2 = 0.95 in the test dataset. Similar results were achieved for canopy size with an estimation accuracy of R2 = 0.93 in the test dataset. The processing time for each image frame of 20 M pixels with each crop row geo-referenced was 2.22 s (including 1.80 s for pre-processing), which was more efficient than traditional mosaic-based image processing methods. An open-source automated image-processing framework was developed for cotton emergence evaluation and is available to the community for efficient data processing and analytics.

[1]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[2]  F. Forcella,et al.  Modeling seedling emergence , 2000 .

[3]  F. Baret,et al.  Estimates of plant density of wheat crops at emergence from very low altitude UAV imagery. , 2017 .

[4]  Rasmus Nyholm Jørgensen,et al.  RoboWeedSupport - Detection of weed locations in leaf occluded cereal crops using a fully convolutional neural network , 2017 .

[5]  Ronald Davis,et al.  Neural networks and deep learning , 2017 .

[6]  Vijay Kumar,et al.  Counting Apples and Oranges With Deep Learning: A Data-Driven Approach , 2017, IEEE Robotics and Automation Letters.

[7]  L. M. Bugayevskiy,et al.  Map Projections: A Reference Manual , 1995 .

[8]  Xin Zhang,et al.  Multi-class fruit-on-plant detection for apple in SNAP system using Faster R-CNN , 2020, Comput. Electron. Agric..

[9]  Surya Kant,et al.  Estimation of crop plant density at early mixed growth stages using UAV imagery , 2019, Plant Methods.

[10]  Richard O. Duda,et al.  Use of the Hough transformation to detect lines and curves in pictures , 1972, CACM.

[11]  Lori J. Wiles,et al.  The cost of counting and identifying weed seeds and seedlings , 1999, Weed Science.

[12]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[13]  S. Sankaran,et al.  High-Resolution Aerial Imaging Based Estimation of Crop Emergence in Potatoes , 2017, American Journal of Potato Research.

[14]  Kenneth A. Sudduth,et al.  Comparison of electromagnetic induction and direct sensing of soil electrical conductivity , 2003 .

[15]  K. Ghassemi-Golezani,et al.  Effects Of Seed Vigor On Growth And Grain Yield Of Maize , 2014 .

[16]  Ruizhi Chen,et al.  Monitoring cotton (Gossypium hirsutum L.) germination using ultrahigh-resolution UAS images , 2018, Precision Agriculture.

[17]  Xin Zhang,et al.  Faster R–CNN–based apple detection in dense-foliage fruiting-wall trees using RGB and depth features for robotic harvesting , 2020 .

[18]  Loris Nanni,et al.  Handcrafted vs. non-handcrafted features for computer vision classification , 2017, Pattern Recognit..

[19]  Urs Schmidhalter,et al.  Digital Counts of Maize Plants by Unmanned Aerial Vehicles (UAVs) , 2017, Remote. Sens..

[20]  Terry Griffin,et al.  Early-Season Stand Count Determination in Corn via Integration of Imagery from Unmanned Aerial Systems (UAS) and Supervised Learning Techniques , 2018, Remote. Sens..

[21]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[22]  Lav R. Khot,et al.  Field-based crop phenotyping: Multispectral aerial imaging for evaluation of winter wheat emergence and spring stand , 2015, Comput. Electron. Agric..

[23]  Henrik Skov Midtiby,et al.  Plant species classification using deep convolutional neural network , 2016 .

[24]  Jianfeng Zhou,et al.  Cotton Yield Estimation from UAV-Based Plant Height , 2019, Transactions of the ASABE.

[25]  Xiuliang Jin,et al.  A method to estimate plant density and plant spacing heterogeneity: application to wheat crops , 2017, Plant Methods.

[26]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[27]  Andreas Kamilaris,et al.  Deep learning in agriculture: A survey , 2018, Comput. Electron. Agric..

[28]  Baskar Ganapathysubramanian,et al.  An explainable deep machine vision framework for plant stress phenotyping , 2018, Proceedings of the National Academy of Sciences.

[29]  Dennis B. Egli,et al.  Seed Vigor and the Uniformity of Emergence of Corn Seedlings , 2012 .

[30]  Jianfeng Zhou,et al.  Evaluation of Cotton Emergence Using UAV-Based Narrow-Band Spectral Imagery with Customized Image Alignment and Stitching Algorithms , 2020, Remote. Sens..

[31]  Marcel Salathé,et al.  Using Deep Learning for Image-Based Plant Disease Detection , 2016, Front. Plant Sci..

[32]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[33]  S. Sankaran,et al.  Low-altitude, high-resolution aerial imaging systems for row and field crop phenotyping: A review , 2015 .

[34]  Chenghai Yang,et al.  Rapeseed Seedling Stand Counting and Seeding Performance Evaluation at Two Early Growth Stages Based on Unmanned Aerial Vehicle Imagery , 2018, Front. Plant Sci..

[35]  Rui Li,et al.  Improved Kiwifruit Detection Using Pre-Trained VGG16 With RGB and NIR Information Fusion , 2020, IEEE Access.

[36]  Maryam Rahnemoonfar,et al.  Deep Count: Fruit Counting Based on Deep Simulated Learning , 2017, Sensors.