Researchers From The Ohio State University Use Machine Learning And Drones To Develop A Novel Method For Determining Health Of Soybean Fields

This Article is written as a summary by Marktechpost Staff based on the paper 'Assessing the efficacy of machine learning techniques to characterize soybean defoliation from unmanned aerial vehicles'. All Credit For This Research Goes To Researchers on This Project. Checkout the paper and post.

Please Don't Forget To Join Our ML Subreddit

Researchers at The Ohio State University have devised a revolutionary approach for detecting crop health using a mix of drones and machine learning algorithms to develop a new tool that might help future farmers.

The study, published in the journal Computers and Electronics in Agriculture, looks at employing neural networks to assist in quantifying crop defoliation or the loss of a plant’s leaves on a large scale. Disease, stress, raising livestock, and, more commonly, bug and other pest infestations can all contribute to the loss of crops.

Complete crop fields can be ruined if left unchecked, decreasing an entire region’s agricultural yield significantly. To tackle this, scientists looked at soybeans, a cash crop considered one of the four staples of world agriculture.

Zichen Zhang, the study’s primary author and a graduate computer science and engineering student at Ohio State, utilized an Unmanned Aerial Vehicle (UAV), or drone, to acquire aerial photos of five soybean fields in Ohio between August and September 2020. After cutting each UAV picture into smaller images, the researchers finally collected over 97,000 shots that could classify either healthy or defoliated.

According to the USDA, the United States is the world’s biggest producer of soybeans and the second-largest exporter. On the other hand, domestic farmers are rushing to meet the demand: To meet customer demand, almost 90 million acres of soybean crops were expected to be sown last year.

As soybeans are a significant source of oil, food, and protein across many parts of the world, a reduction in soybean output in the United States might have far-reaching implications. However, Zhang’s research, one of the first to use non-invasive methods to define crop health, can aid in determining the possibility of a production decline due to defoliation.

After manually sorting through the photos, researchers discovered that over 67,000 could be classified as healthy, while nearly 30,000 exhibited symptoms of defoliation, a ratio of more than 2-to-1. The researchers then utilized this data set to examine the capabilities of various learning algorithms to detect which crops were defoliated and avoid generating false conclusions about healthy soybean crops.

However, after finding that none of the existing learning classifiers could provide the level of precision they desired, the researchers opted to build their own deep learning model from the ground up. Defonet is the end product, a neural network capable of studying and accurately answering the study’s initial defoliation questions.

According to research co-author Christopher Stewart, an associate professor of computer science and engineering, Defonet might alter the agriculture industry’s decision-making process in coping with significant crop losses if implemented in the field.

Stewart added that in the coming years, they’d have to significantly boost food production to keep up with demand. The concept behind digital agriculture is to use machine learning and other technology to ensure that each seed sown is cultivated as efficiently as possible.

The study was supported by the National Science Foundation and Sami Khanal, an assistant professor of food, agricultural, and biomedical engineering, Amy Raudenbush, a research associate in entomology, and Kelley Tilmon, an associate professor of entomology, were also co-authors on the paper.