close
Wednesday December 25, 2024

AI pest detection tool developed to promote production

By Nadeem Shah
July 16, 2024
In this picture taken on February 23, 2020, officials of the Agriculture Department on a tractor spray pesticides to kill locusts as a farmer works in a field in Pipli Pahar village in Punjab. — AFP
In this picture taken on February 23, 2020, officials of the Agriculture Department on a tractor spray pesticides to kill locusts as a farmer works in a field in Pipli Pahar village in Punjab. — AFP

MULTAN: The innovative tools of Artificial Intelligence (AI) have become highly successful in smart agriculture, leading to increased production. A real-life example of AI automating pest detection is the use of AI pest detection and pheromone traps by cotton growers to manage bollworms.

This system helps farmers determine when and how much pesticide is required to avoid over-spraying and achieve higher yields. The potential of data science, drone technology, and other smart agriculture innovations are also successful tools of AI in boosting production while minimizing costs. Currently, drone technologies are being utilized for spraying in Punjab’s villages, to increase productivity, reduce production costs, and improve the health outcomes for farmers. The South Punjab Agriculture Department (SPAD) is also working diligently to develop new AI tools to enhance pest management and increase production.

As part of a series to promote AI tools in South Punjab’s smart agriculture, SPAD has launched a ‘Virtual Cotton Pest Management Hub’. Under the leadership of South Punjab Secretary of Agriculture Saqib Ali Attil, an online Google workbook called “Virtual Cotton Pest Management Hub” has been developed and launched.

This workbook allows for the online tracking of cotton pest scouting, management, and other related activities. The workbook consists of three separate sheets, each named according to the type of data from the Department of Agriculture Extension, and Pest Warning Monitoring Team, and color-coded accordingly. The Google Sheet developed by SPAD is aimed at real-time monitoring of cotton data. Only authorized personnel will have access to enter data in this Google Sheet.

The objective of developing this Google sheet is to ensure real-time monitoring of cotton data. Mr. Atil further explained that all other field activities, including nutrition and pest scouting, recorded in the workbook will be monitored. He added that 11 entomologists from various departments of the South Punjab Department of Agriculture have been assigned duties to monitor cotton pests and cross-verify the data recorded by the Pest Warning and Extension Department. Each monitoring officer is responsible for monitoring at least 70 hotspots per month. The observations made by the monitors will be recorded on the third sheet of the Virtual Cotton Pest Management Hub during their visits.

The hotspots and treatment status will be recorded on the Google Sheet by the Department of Agriculture Extension and Pest Warning staff. The monitoring officers will also be responsible for verifying the advisory. They can also check the status of the cotton crop in their designated areas, and monitor the use of pesticides, nozzles, the type of machinery used for spraying, and spray techniques.

Further, they can provide remarks in their respective Google Sheets. The objective of developing this online mechanism is to achieve better production by timely controlling cotton pests using modern technology, he emphasized. Agriculture scientists have also established an AI laboratory in collaboration with China in Pakistan to promote AI Smart Agriculture and boost farming in the province. The lab aims to promote state-of-the-art agricultural practices to enhance field productivity. AI can assist breeders in making decisions by providing insights based on data. For example, it can recommend which plants to crossbreed to achieve desired traits or identify the ideal environmental conditions for specific varieties. It is important to mention here that AI refers to the simulation of human intelligence in machines, enabling them to perform tasks that typically require human intelligence. These tasks include learning from experience, problem-solving, pattern recognition, understanding natural language, and making decisions. Machine Learning (ML) develops algorithms that learn to perform specific tasks based on a given data set. It is a subfield of artificial intelligence extensively used in research and industry. Supervised learning tasks aim to predict an output (either a discrete label for classification or a numerical value for regression) for a given object, based on a set of input features that describe the object. Supervised methods utilize labelled input data. Unsupervised methods, on the other hand, do not use labels but identify groups or trends in data, as observed by agricultural scientists. Plant breeding is the science and art of selecting and crossing plants with desirable traits to develop plant varieties that are better suited for specific purposes, such as improving crop yields, disease resistance, tolerance to environmental stress, and overall plant quality. AI refers to the simulation of human intelligence in machines, enabling them to perform tasks that typically require human intelligence, they added.