Nafaa Ben Hedi Jabeur

We propose an intelligent system by integrating of Infrared Thermography (IRT), Machine learning (ML) and Deep
learning (DL) techniques to create an innovative diagnostic tool for the early diagnosis of Mastitis. Infrared
thermography allows for real-time, contact-less, continuous monitoring of udder temperature variations, which is
found to be one of the key indicators of mastitis along with ocular surface temperature. Deep learning algorithms
will be trained on extensive datasets encompassing thermal patterns associated with healthy and infected udders
to identify the infection caused by mastitis at the early stage of disease. The resulting model will enable automated
identification of mastitis at its early stages with better accuracy, distinguishing between clinical and sub-clinical
cases and give the alert message to the farmers and farm owners to take immediate medical attention to control
the disease severity of the infected cow and suppress the spreading of mastitis to other cattle. Our interdisciplinary
team, comprising experts in veterinary medicine, agriculture, computer science, and data analytics along with
university students who will collaborate to build and validate the system across diverse dairy farm environments.
Field trials will assess the tool's efficacy in various climates, herd sizes, and management practices, ensuring its
adaptability to real-world conditions.
This project aligns with the objectives of innovating for a sustainable agriculture and improving the livelihoods of
farmers. The proposed solution holds the potential to enhance and improve mastitis management, contributing to
increased dairy productivity, reduced antibiotic usage, and enhanced animal welfare by focusing on providing
quality food to the society.

The adoption of drone related technologies in a wide range of business applications is increasing exponentially. This adoption is particularly motivated by the capability of drones to provide professionals, researchers, and practitioners with a bird-view on objects and events of interest as well as to allow them reach areas with limited or costly access. In the specific context of supply chain, drones have been recognized as promising, cost-effective tools in last-mile delivery, where the focus is on transporting parcels from a distribution hub to their final destinations within predefined short deadlines. Although several professional multi-copter drones are already available, their limited flight-times still present a major challenge toward delivering merchandise to clients within satisfactory time windows. In order to address this shortcoming, recent studies and initiatives are investigating hydrogen-fuelled systems as a serious alternative to the current electrochemical cells (batteries) for their high energy density, zero-emission, and low maintenance costs. In this proposal, we are aiming to explore the potential of powering multi-copter drones with hydrogen for the important field of last-mile parcel delivery in Oman. We are also aiming to explore design and structural options that would make these drones more robust as they are not yet capable of operating correctly in temperatures higher than 40 °C.

SNEPMURe stands for Sensor Networks for an Enhanced and Personalized Management and Use of Resources. It aims at elaborating mechanisms, protocols, and algorithms allowing static and mobile sensors to collaborate within the same Wireless Sensor Network (WSN). WSNs are promising tools in solving some of the most pressing challenges facing the world today, including the need for conserving energy, carrying out sustainable development, lowering the cost of healthcare, and improving governmental response to natural disasters. In the context of Oman, these emergent tools could be very beneficial in the monitoring of the existing rich and abundant natural resources.

Basically, a WSN consists of hundreds of sophisticated small wireless devices (sensors) that could be deployed over a geographic space to cooperatively monitor physical or environmental conditions, such as temperature, sound, vibration, pressure, motion, and pollutants. Since the energy and processing constrained sensors are frequently prone to failure, robbery, and destruction, the WSN commonly faces problems in managing its resources as well as the resources related to its application.

By increasing the context awareness of sensors and allowing them to make a convenient use of data and environmental semantics, these small devices will leverage their interactions in order to adapt their processing with respect to current user requirements and the intended use of the WSN. The ultimate goal of this interaction is to enhance and personalize the management and use of the available resources.

The contributions of SNEPMURe will include, but not limited to:
- Personalize the WSN services with respect to the user requirements.
- Personalize the WSN processing for better interactions between sensors.
- Build a prototype that complies with the objectives of SNEPMURe.

The project aims to survey the existing positioning technologies being used for marine applications. It also aims to study their effectiveness under several environmental conditions. Based on the results of the study, a framework for a new marine positioning engine will be proposed.

The projects aims to create an automated sensor web system for telemetry applications. Since a "deploy-and-ignore" approach does not work, especially in harsh and remote areas, sensors must be endowed with intelligent mechanisms to increase their context-awareness, set up and maintain safe communications pathways, and report the right data, at the right time, to the right sink, while preserving their limited energies.