About

Prof. Faouzi Alaya Cheikh is the head of the Visual Information Processing Lab at NTNU’s Gjøvik campus. He brings extensive expertise in image and signal processing, computer vision, the Internet of Things (IoT), and advanced artificial intelligence algorithms.

The PheNo-NTNU team comprises specialists in spectral imaging, colour science, AI, and high-performance computing. Their involvement underscores NTNU’s strong commitment to translating cutting-edge research into practical solutions for real-world challenges in plant phenotyping and environmental monitoring.

With its top experts in deep learning, image processing and analysis, and IoT sensor networks, NTNU will support PheNo infrastructure users in adopting IoT systems for data collection, as well as in training and deploying deep learning models for analyzing the collected data.

NTNU will also test and maintain a library of deep learning models tailored to typical phenotyping needs, which will be made available to infrastructure users. The central hub for data storage, analysis, and computing in the project will be SIGMA2, with usage coordinated by NTNU lab engineers.

The team will focus on designing an optimal IoT network architecture to enable efficient data collection across multiple field sites. Their work also includes the testing and calibration of various sensor types to ensure consistent, high-quality data acquisition. A core aspect of their contribution will be the development of advanced multimodal data processing and analysis algorithms, enabling the integration of heterogeneous datasets for deeper insights into plant phenotyping.

Furthermore, the team will explore the application of deep learning techniques to extract phenotypic traits from complex sensor data and to model their relationships with key growth factors. By leveraging high-resolution multispectral and hyperspectral imaging, combined with state-of-the-art AI methods, the PheNo-NMBU team aims to deliver innovative, data-driven solutions that support sustainable agriculture

Upcoming Services

• To develop an ideal scalable IoT network architecture for simultaneously collecting sensor data from different fields and plants
• To develop and implement machine learning algorithms for automatically processing and analysing multi-sensor data (time series, images, point clouds and multispectral images) to extract/record plants and seeds phenotypes
• To perform multimodal data fusion and understanding based on deep learning (DL)
• To develop DL algorithms for understanding the latent variables from the observable variables, thus connecting the plant and environment conditions to phenotypes and growth factors