Predicting cotton yield of small field plots in a cotton breeding program using UAV imagery data

Maja, Joe Mari; Campbell, B. Todd; Neto, J. Camargo; Astillo, Philip Virgil

doi:10.1117/12.2228929

Cited by 10 publications

(11 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In their study, Paproki, Sirault, Berr, Furbank, and Fripp (2012) created a 3D point cloud out of multiple images of a single cotton plant to visualize its organs, and Yeom, Jung, Chang, Maeda, and Landivar (2018) used a DJI Phantom 4, with integrated RGB sensor, to collect unmanned aerial vehicle (UAV) open boll detection and yield estimation data. In another study Maja et al (2016), predicted cotton yield using UAV imagery, and Jung et al (2018) used an unmanned aerial system (UAS)-assisted framework for the selection of highyielding cotton genotypes. Most of the work on open-boll detection and yield estimation has focused on using UAV systems and not much has been documented on the use of proximal sensors.…”

Section: Core Ideasmentioning

confidence: 99%

See 1 more Smart Citation

Cotton boll distribution and yield estimation using three‐dimensional point cloud data

et al. 2020

View full text Add to dashboard Cite

Node-by-node boll mapping has been used to determine the effects of several environmental and management strategies, including irrigation rate and cultivar, on cotton (Gossypium hirsutum L.) boll distribution. The advent of consumer 3D digital imaging may make rapid node-by-node mapping of greater acreage possible in the future. Effects of irrigation rate and cultivar on boll distribution were determined from line plots of node-specific boll distribution and boll accumulation estimate data and from vertical box and whisker plots of boll fraction using data collected by a 3D sensor system to rapidly detect open bolls before harvest over three seasons. Differences were observed between the dryland (2018 only), low irrigation, and high irrigation rates in terms of boll fraction and boll accumulation for each cultivar. The low irrigation tended to produce bolls more toward the bottom of the plant, while the high irrigation produced bolls towards the top of the plant. Yield correlation between sensor obtained and manually counted measurements was strong, with r 2 values as high as 0.87. Results obtained indicated that differences among irrigation rate and cultivar were identifiable using the sensor system during the 3 yr research study. Similar systems may allow rapid, broad-scale identification of boll distribution and yield in breeding and physiology research in the future, leading to improved identification of elite cultivars and improved management practices in cotton. Abbreviations: RGB, red-green-blue; UAV, unmanned aerial vehicle This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

show abstract

Section: Core Ideasmentioning

confidence: 99%

“…In another study Maja et al. (2016), predicted cotton yield using UAV imagery, and Jung et al. (2018) used an unmanned aerial system (UAS)‐assisted framework for the selection of high‐yielding cotton genotypes.…”

Section: Introductionmentioning

confidence: 99%

Cotton boll distribution and yield estimation using three‐dimensional point cloud data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Small unmanned aircraft systems (sUAS) are quickly evolving into a useful platform for a variety of agricultural tasks including detecting diseases and weeds [2,3], yield prediction [4], water stress [5] and spraying chemicals [6]. Images collected by sUAS has been used to validate models using statistical analysis [5] and most recently using artificial intelligence [7].…”

Section: Introductionmentioning

confidence: 99%

RFID and Drones: The Next Generation of Plant Inventory

et al. 2021

View full text Add to dashboard Cite

Collection of plant inventory (i.e., count, grade, plant size, yield) data is time-consuming, costly, and can be inaccurate. In response to increasing labor costs and shortages, there is an increased need for the adoption of more automated technologies by the nursery industry. Growers, small and large, are beginning to adopt technologies (e.g., plant spacing robots) that automate or augment certain operations, but greater strides must be taken to integrate next-generation technologies into these challenging unstructured agricultural environments. The main objective of this work is to demonstrate merging specific ground and aerial-based technologies (Radio Frequency Identification (RFID), and small Unmanned Aircraft System (sUAS)) into a holistic systems approach to address the specific need of moving toward automated on-demand plant inventory. This preliminary work focuses on evaluating different RFID tags with respect to their distance and orientation to the RFID reader. Fourteen different RFID tags, five distances (1.5 m, 3.0 m, 4.5 m, 6.0 m, and 7.6 m), and four tag orientations (the front of the tag (UP), back of the tag (DN), tag at sideways left (SL), and tag at sideways right (SR)) were assessed. Results showed that the tag upward orientation resulted in the highest scanning total for both the laboratory and field experiments. Two orientations (UP and SR) had significant effect on the scan total of tags. The distance between the reader and the tags at 1.5 m and 6.0 m did not significantly affect the scanning efficiency of the RFID system in horizontally fixed (p-value > 0.05) position regardless of tags. Different tag designs also produced different scan totals. Overall, since most of the tags were scanned at least once (except for Tag 6F), it is a very promising technology for use in nursery inventory data acquisition. This work will create a unique inventory system for agriculture where locations of plants or animals will not present a barrier as the system can easily be mounted on a drone. Although these experiments are focused on inventory in plant nurseries, results for this work has potential for inventory management in other agricultural sectors.

show abstract

“…sUASs allow farmers to quickly survey large plots of land using aerial imagery. sUAS imagery has been used to detect diseases and weeds [14,15], predict cotton yield [16], measure the degree of stink bug aggregation [17], and identify water stress in ornamental plants [18]. Several thermal and spectral indices have been correlated to biophysical plant parameters based on sUAS imagery [19,20].…”

Section: Introductionmentioning

confidence: 99%

Watson on the Farm: Using Cloud-Based Artificial Intelligence to Identify Early Indicators of Water Stress

et al. 2019

View full text Add to dashboard Cite

As demand for freshwater increases while supply remains stagnant, the critical need for sustainable water use in agriculture has led the EPA Strategic Plan to call for new technologies that can optimize water allocation in real-time. This work assesses the use of cloud-based artificial intelligence to detect early indicators of water stress across six container-grown ornamental shrub species. Near-infrared images were previously collected with modified Canon and MAPIR Survey II cameras deployed via a small unmanned aircraft system (sUAS) at an altitude of 30 meters. Cropped images of plants in no, low-, and high-water stress conditions were split into four-fold cross-validation sets and used to train models through IBM Watson’s Visual Recognition service. Despite constraints such as small sample size (36 plants, 150 images) and low image resolution (150 pixels by 150 pixels per plant), Watson generated models were able to detect indicators of stress after 48 hours of water deprivation with a significant to marginally significant degree of separation in four out of five species tested (p < 0.10). Two models were also able to detect indicators of water stress after only 24 hours, with models trained on images of as few as eight water-stressed Buddleia plants achieving an average area under the curve (AUC) of 0.9884 across four folds. Ease of pre-processing, minimal amount of training data required, and outsourced computation make cloud-based artificial intelligence services such as IBM Watson Visual Recognition an attractive tool for agriculture analytics. Cloud-based artificial intelligence can be combined with technologies such as sUAS and spectral imaging to help crop producers identify deficient irrigation strategies and intervene before crop value is diminished. When brought to scale, frameworks such as these can drive responsive irrigation systems that monitor crop status in real-time and maximize sustainable water use.

show abstract

Predicting cotton yield of small field plots in a cotton breeding program using UAV imagery data

Cited by 10 publications

References 4 publications

Cotton boll distribution and yield estimation using three‐dimensional point cloud data

Cotton boll distribution and yield estimation using three‐dimensional point cloud data

RFID and Drones: The Next Generation of Plant Inventory

Watson on the Farm: Using Cloud-Based Artificial Intelligence to Identify Early Indicators of Water Stress

Contact Info

Product

Resources

About