IAGRI 4.0

IAGRI 4.0

IAgri 4.0 Project, Artificial Intelligence for early detection of crop diseases and pests

• iAgri 4.0 designs a device that combines image analysis and artificial intelligence to measure the activity and health of crops in their early stages

• onTech Innovation coordinates this initiative, funded by the Ministry of Industry, with the participation of Solutia Innovaworld Technologies, Biopharma Research of the Econatur group and the University of Seville

The iAgri 4.0 project has developed a device that facilitates early detection of diseases and pests in crops. The project has counted with the participation of Solutia Innovaworld Technologies, Biopharma Research of the Econatur group and the University of Seville. The onTech Innovation Cluster has been in charge of coordinating this initiative funded by the Ministry of Industry, Tourism and Trade, within the last call for grants for Innovative Business Groups (AEIs).

With this initiative, onTech Innovation consolidates its capacity to promote innovation in Con Andalucía with the launch of this new iAgri 4.0 project, in addition to two others underway, which together mobilise an investment of almost 1.4 million euros and which are providing technological solutions in the areas of cybersecurity, agriculture and waste treatment.

Crop diseases and pests are some of the most worrying problems for farmers, as a large part of the annual production can be lost due to them. To mitigate their impact, it is necessary to detect them at an early stage. This project aims to detect crop diseases using image analysis techniques and artificial intelligence, which also aims to make an early diagnosis of nutritional deficiencies in plants.

The main objective of the device designed by iAgri 4.0 is to be able to detect the temperature of the leaves in order to monitor both the activity and the health of the plant. To this end, it has been equipped with an infrared detector and a conventional camera. In addition, it has a thermal blanket and thermocouples as elements for calibrating the camera.

One of the methods used to measure the activity and health of crops is based on measuring the temperature of plants, which is related to how they are carrying out stomatal transpiration. Plants open their stomata, exchanging gases between leaves and atmosphere as a cooling mechanism and to control their temperature.

Low-cost, easy-to-use thermal camera

During periods of water stress, such as drought, the plant closes its stomata to reduce transpiration and prevent water loss. This is automatically reflected in an increase in temperature. In order to be able to detect this and similar effects, a low-cost, easy-to-use thermal camera has been developed. The aforementioned thermal blanket of this device provides a known reference temperature, as it is heated by adjustable thermocouples, to calibrate the chamber.

Artificial Intelligence to detect pests and diseases

In addition, an artificial intelligence system has also been developed to detect the presence of pests and diseases in crops from images taken with a mobile phone camera. By training with images of affected crops and healthy crops, the artificial intelligence learns to distinguish one from the other. In addition, the tools used are intended to automate and increase the frequency of analysis, thus increasing the possibility of determining and predicting crop health and potential biotic risks.

An app to evaluate images immediately

For the development of this artificial intelligence system, the project has relied on Econatur’s experimental farm (La Añoreta) equipped with sensors, both atmospheric and edaphic. These sensors allow data to be taken on temperature, humidity, conductivity, and the control of irrigation provided to the crop, among others. This allows control of the inputs and water provided to the crop, controlling the water footprint and avoiding the excess of fertilisers in the soil, allowing sustainable crop management without losing profitability.

This analytical development is supported by two technological enablers: an autonomous ground vehicle and the functionalities integrated in iAgri’s web platform and mobile app (image management and inspection route configuration). The vehicle, in addition to having an IMU to allow fully autonomous navigation, has two mobile supports where the visual camera and the thermal camera are anchored.

Through the inspection route configuration section, the user can define the topography and configuration of the farm to be inspected, including options for spot or periodic inspection. The app will be used both to download the route to the vehicle and to receive the collected samples. This data will be available on the web platform so that the developed algorithms can detect and notify the user in an automated way when and in which sectors of the farm any incidence related to pests or water stress situations was detected. Alternatively, the user can directly take images from his smartphone and evaluate them in situ via the app.

CHARACTERISTICS OF THE IAGRI 4.0 PROJECT

Title: Agronomic analysis based on machine vision and thermal distribution techniques to determine and predict crop quality and the possible occurrence of biotic risks

Summary: Development of a system based on artificial intelligence for the diagnosis of nutritional deficiencies, pests and diseases; enabling decision making and sustainable crop management

Reference: AEI-010500-2021-13

Project start date: 08/04/2021 Project end date: 31/03/2022

Financing: 317,823 €