Embrapa Instrumentation (SP) has developed a method that uses image sensors and artificial intelligence to identify the fall armyworm, one of the most aggressive pests of corn crops. The system analyzes digital images and identifies the caterpillar both on the plant's leaf and on the corn cob. In this way, it is possible to minimize the subjectivity of traditional methods, which are laborious and depend on human observation to perform the same task.<\/p>\n
Corn is one of the most widely cultivated cereals in the world, and the fall armyworm (Spodoptera frugiperda) is one of the main pests of the crop. This insect can cause losses capable of compromising 70% of production, according to researchers at Embrapa. The caterpillar attacks plants in both the vegetative and reproductive phases.<\/p>\n
Alternative method<\/strong><\/h4>\nThe methodology was published in the journal Electronics, in the article "Computational Intelligence Approach for Fall Armyworm Control in Maize Crop," by Alex Bertolla and Paulo Cruvinel. They explain that the work was motivated by the discrepancy between the current detection method and the intended result, which led them to research an alternative for the early detection of pests in cultivated areas.<\/p>\n
![](\"https:\/\/i0.wp.com\/opresenterural.com.br\/wp-content\/uploads\/2025\/09\/Foto-Joana-Silva.jpg?resize=300%2C214&ssl=1\")
<\/p>\nPhoto: Joana Silva<\/p>\n<\/div>\n
Bertolla explains that the study focused on recognizing and classifying dynamic patterns of the fall armyworm, which, in addition to corn, attacks various agricultural crops such as soybeans and cotton. He emphasizes that the solution obtained can help agronomists and laboratories achieve more precise results.<\/p>\n
To facilitate image capture, a simple camera can be attached to agricultural implements to collect images of caterpillars on corn plants, both leaves and ears, while performing field operations. The camera does not need to be expensive; it only needs to produce images with good resolution.<\/p>\n
Process integration<\/strong><\/h4>\nThe method integrates digital image and signal processing, multivariate statistics, machine learning techniques, and computer vision. \u201cMachine learning describes the ability of systems to learn from customized training data to automate and solve associated tasks. In conjunction with this concept, deep learning is also a machine learning process, but based on artificial neural networks with a set of purpose-specific layers,\u201d explains Cruvinel.<\/p>\n
Bertolla explains that the computational algorithm has the ability to evaluate digital images of different growth stages of the caterpillar located on corn plants, as well as its developmental stage and frequency of occurrence in the crop area. The program was developed in Python, a high-level programming language widely used in data science and machine learning.<\/p>\n
The study evaluated a total of 2,280 images of fall armyworms present on leaves or ears of corn in the corn growing area, in order to classify five distinct stages of pest development during the production cycle.<\/p>\n
Basis for future studies<\/strong><\/h4>\nCruvinel and Bertolla report that the quality of the results confirmed the recommendation to use the deep learning-based framework. The data underwent analysis of accuracy, precision, processing time, and hardware performance for the proposed scenario.<\/p>\n
\u201cThe method also showed considerable results in terms of hardware performance and processing time, which may also be interesting for future action related to making this technology available in an embedded version, for direct use as part of agricultural implements,\u201d points out Cruvinel.<\/p>\n
They suggest, for future work, including other artificial intelligence techniques dedicated to pest control for pattern recognition and classification, both in an unsupervised manner and through the use of a multispectral camera mounted on an Unmanned Aerial Vehicle (UAV) for real-time operation.<\/p>\n<\/div>\n
THE MM Cereais<\/strong> works with the best grains on the market and also keeps you up to date with the latest news and analyses on agribusiness.
\nDon't forget to follow our social networks.<\/em><\/h6>\n
<\/p>\nPhoto: Joana Silva<\/p>\n<\/div>\n
Bertolla explains that the study focused on recognizing and classifying dynamic patterns of the fall armyworm, which, in addition to corn, attacks various agricultural crops such as soybeans and cotton. He emphasizes that the solution obtained can help agronomists and laboratories achieve more precise results.<\/p>\n
To facilitate image capture, a simple camera can be attached to agricultural implements to collect images of caterpillars on corn plants, both leaves and ears, while performing field operations. The camera does not need to be expensive; it only needs to produce images with good resolution.<\/p>\n
Process integration<\/strong><\/h4>\nThe method integrates digital image and signal processing, multivariate statistics, machine learning techniques, and computer vision. \u201cMachine learning describes the ability of systems to learn from customized training data to automate and solve associated tasks. In conjunction with this concept, deep learning is also a machine learning process, but based on artificial neural networks with a set of purpose-specific layers,\u201d explains Cruvinel.<\/p>\n
Bertolla explains that the computational algorithm has the ability to evaluate digital images of different growth stages of the caterpillar located on corn plants, as well as its developmental stage and frequency of occurrence in the crop area. The program was developed in Python, a high-level programming language widely used in data science and machine learning.<\/p>\n
The study evaluated a total of 2,280 images of fall armyworms present on leaves or ears of corn in the corn growing area, in order to classify five distinct stages of pest development during the production cycle.<\/p>\n
Basis for future studies<\/strong><\/h4>\nCruvinel and Bertolla report that the quality of the results confirmed the recommendation to use the deep learning-based framework. The data underwent analysis of accuracy, precision, processing time, and hardware performance for the proposed scenario.<\/p>\n
\u201cThe method also showed considerable results in terms of hardware performance and processing time, which may also be interesting for future action related to making this technology available in an embedded version, for direct use as part of agricultural implements,\u201d points out Cruvinel.<\/p>\n
They suggest, for future work, including other artificial intelligence techniques dedicated to pest control for pattern recognition and classification, both in an unsupervised manner and through the use of a multispectral camera mounted on an Unmanned Aerial Vehicle (UAV) for real-time operation.<\/p>\n<\/div>\n
THE MM Cereais<\/strong> works with the best grains on the market and also keeps you up to date with the latest news and analyses on agribusiness.
\nDon't forget to follow our social networks.<\/em><\/h6>\n
Cruvinel and Bertolla report that the quality of the results confirmed the recommendation to use the deep learning-based framework. The data underwent analysis of accuracy, precision, processing time, and hardware performance for the proposed scenario.<\/p>\n
\u201cThe method also showed considerable results in terms of hardware performance and processing time, which may also be interesting for future action related to making this technology available in an embedded version, for direct use as part of agricultural implements,\u201d points out Cruvinel.<\/p>\n
They suggest, for future work, including other artificial intelligence techniques dedicated to pest control for pattern recognition and classification, both in an unsupervised manner and through the use of a multispectral camera mounted on an Unmanned Aerial Vehicle (UAV) for real-time operation.<\/p>\n<\/div>\n