Robust Embedded AI for Low-Cost Infrared Spectral Recognition: Tools for Ecology, Agriculture, and Recycling

About the Project

Infrared spectral recognition is a powerful way to identify plants, assess ecological conditions, and distinguish between different materials for recycling. However, most current systems are heavy, expensive, and unsuitable for widespread use in the field. This project aims to change that by developing low-cost, lightweight devices supported by embedded artificial intelligence. By drawing on advances in machine learning, single-board computers and compact sensors, it will contribute directly to the growing field of computational sustainability. The results could transform how we monitor biodiversity, manage crops, and improve recycling processes using affordable, accessible technology.

There is increasing global demand for sustainable solutions that balance affordability, accuracy, and environmental responsibility. Farmers need timely insights to cope with crop stress caused by climate change. Ecologists require new tools to measure biodiversity across large areas. Recycling industries face pressure to improve material sorting and reuse. At the same time, debates in technology and society highlight the need for low-energy embedded computing rather than relying entirely on cloud systems. This research connects to those discussions by developing both portable devices and systems that can be mounted on drones, offering new flexibility in monitoring landscapes and managing resources.

The project will develop operational prototypes that focus on three interconnected challenges:

• Improving signal quality from affordable infrared sensors through innovative noise reduction methods.
• Ensuring reliability by creating practical calibration approaches and contributing to shared spectral reference libraries.
• Deploying efficient AI models so that recognition can occur directly in the field without dependence on expensive hardware.

This project will generate new methods for embedding artificial intelligence into low-cost spectral recognition systems. Its outcomes will include practical software frameworks, calibration protocols, and a working prototype that demonstrates value across different domains. The broader impact lies in supporting biodiversity research, strengthening food security, and enabling more effective recycling practices. For a motivated doctoral student, the project offers the chance to develop cutting-edge technical skills while working at the frontier of sustainability and applied AI, with opportunities for fieldwork in diverse environments.

Supervisors: Dr Brian Davison, Prof Chan See

Applications accepted all year round

Self-Funded PhD Students Only

Academic qualifications

Have, or expect to achieve by the time of start of the studentship a first-class honours degree, or a distinction at master level, ideally in Artificial Intelligence, Computer Science with a good functional knowledge of Image processing, Sensor-based systems, Internet of Things.

English language requirement

IELTS score must be at least 6.5 (with not less than 6.0 in each of the four components). Other, equivalent qualifications will be accepted.

Essential attributes:

• Good understanding of the physics of spectral image processing
• Excellent problem-solving and creative thinking
• Advanced software engineering skills

Desirable attributes:

• Good level of written academic English
• Good understanding of the use and limitations of AI tools for software development
• Experience in evaluating the quality of output from sensor-based systems

APPLICATION CHECKLIST

• Completed application form
• CV
• 2 academic references, using the Postgraduate Educational Reference Form (download)
• Research project outline of 2 pages (list of references excluded). The outline may provide details about:
  1. Background and motivation of the project. The motivation, explaining the importance of the project, should be supported also by relevant literature. You can also discuss the applications you expect for the project results.
  2. Research questions or objectives.
  3. Methodology: types of data to be used, approach to data collection, and data analysis methods.
  4. List of references.

The outline must be created solely by the applicant. Supervisors can only offer general discussions about the project idea without providing any additional support.

• Statement no longer than 1 page describing your motivations and fit with the project.
• Evidence of proficiency in English (if appropriate)

To be considered, the application must use

• the advertised title as project title

For informal enquiries about this PhD project, please contact Dr Brian Davison - B.Davison@napier.ac.uk