CBL - Campus del Baix Llobregat

Projecte llegit

Títol: CWSI (Crop Water Stress Index) monitoring with thermal imaging

Estudiant que ha llegit aquest projecte:

Tutor/a o Cotutor/a: SALCEDO CIDONCHA, RAMON

Departament: DEAB

Títol: CWSI (Crop Water Stress Index) monitoring with thermal imaging

Data inici oferta: 07-01-2025      Data finalització oferta: 07-09-2025

Estudis d'assignació del projecte:
    MU MKET4FOOD+BIO
    MU TECH4AGRI+FOOD

Lloc de realització:

UPC      Departament/centre: EEABB-CDEI

Segon tutor/a (UPC): CABALLERO FLORES, DAVID

Paraules clau:
Robotics, irrigation management, thermal sensors, precision agriculture

Descripció del contingut i pla d'activitats:
Introduction:
The CDEI-UPC is a technological innovation center at the Universitat Politècnica de Catalunya (UPC), specializing in machinery engineering, applied research, and technology transfer. Its main research lines focus on mobile robotics, with a particular emphasis on agricultural and construction applications. Furthermore, EEABB is one of the UPC schools dedicated to promoting education and innovation in agri-food systems and environmental sustainability. The shared interest in applying robotics and technology to agricultural challenges offers a unique opportunity to create synergies between the two centres to advance precision agriculture and sustainable agriculture.

Project description and objectives:
CDEI has developed an innovative system that combines computer vision from an RGB camera with thermal imaging from an infrared camera. This device is integrated into an autonomous ground mobile robot that navigates agricultural fields, collecting plant data and sending it to the cloud to create a digital twin of the field.
The primary objective of the project is to estimate the leaf temperature of crops. Using this data along with ambient temperature, the system calculates water stress indices such as the Crop Water Stress Index (CWSI) to predict the plants' water needs and optimize irrigation management. This work will focus on validating the parameters collected by the system and correlating them with the actual condition of the plants. The crops of interest include lettuce, fava beans, brassicas, tomatoes, and fruit trees.

Unique Opportunity:
This project offers a unique chance to work on a real-world project tackling pressing challenges in precision agriculture. It combines theoretical development with practical application, including experimental validation in real crops. You will join a dynamic and multidisciplinary team of professors, engineers, and students specializing in robotics and agriculture, gaining hands-on experience in cutting-edge research and technology development.

Overview (resum en anglès): Water stress is one of the main limiting factors in horticultural and legume production, affecting crop growth and quality. Its detection is essential to optimize irrigation management and ensure sustainable water use, especially in the Mediterranean region, where temperatures are increasingly higher and rainfall less frequent. In this context, optical sensor technologies offer a non-invasive, rapid, and cost-effective method to assess the water status of plants. This study evaluated the capability of two optical sensors for monitoring water stress in common bean crops under controlled greenhouse conditions. The plants were arranged in 30 paired rows and subjected to two irrigation treatments: normal irrigation and deficit irrigation. Two technologies were used: an infrared camera, which detects the natural radiation emitted by the plant, and a multispectral sensor to measure leaf reflectance. The monitoring system was integrated into a data acquisition platform. The thermal camera was used to record apparent leaf temperature and calculate the Crop Water Stress Index (CWSI) based on the normalized difference between leaf and air temperature, while the optical sensor measured reflectance in the red and near-infrared regions to estimate the Normalized Difference Vegetation Index (NDVI). The evolution of water stress was recorded and analyzed over two measurement sessions, maintaining constant distance and angle relative to the plant canopy. Results showed that CWSI increased significantly under water deficit (p<0.001), demonstrating early sensitivity to changes in water status, while NDVI exhibited a more pronounced decline in advanced stages of stress. Both indices presented a weak inverse correlation (p=0.023). Based on the analysis, it was concluded that thermal-spectral integration can be an effective tool for detecting and monitoring water stress in herbaceous crops. This combination could provide methodological support for future implementation in robotic precision-agriculture platforms. Future work will focus on other plant types or on field-scale validation.

Aquest projecte està relacionat amb l'adaptació al Canvi Climatic? Sí

Aquest projecte està relacionat amb la digitalització del seu àmbit de treball? Sí

© CBLTIC Campus del Baix Llobregat - UPC