Computer vision and image analysis

Incorporating machine vision, pattern recognition, image analysis and eye-tracking, our community of researchers have strong and multifaceted interests in vision. We are developing vision-based solutions for a wide variety of problems from object tracking and navigation to privacy-aware data collection.

Computer vision, machine learning, image and video processing, and applied mathematics are not only academically exciting areas of research, but they are also in high demand in many commercial and industrial applications such as video analytics, robotics, healthcare, autonomous systems, and biometrics, just to name a few. Our current projects are focused on civil infrastructure inspection, analysis and monitoring, privacy-aware surveillance systems, use of emotional responses to improve the quality of human experience and health-related data translation from video imagery. We have strong collaborations with schools of life sciences and medicine and with key industry partners.

Privacy-aware activity recognition

Our research focuses on addressing the challenges of anonymising visual data dynamically to ensure privacy and security by design. We combine Privacy-Enhancing Technologies (PETs) and Security-Enhancing Technologies (SETs) to mitigate the risks associated with decrypting data to address potential vulnerabilities.

recognition

The solutions we are developing implement robust security measures throughout the entire data lifecycle. This includes training machine learning models on anonymised visual data using selective encryption, ensuring end-to-end encryption to protect data at rest, during processing, and in transit. By encrypting the region of interest in visual imagery at the source, our approach allows machine learning and deep learning models to use the data directly for training purposes. This ensures that models focus on abstract visual information rather than private details such as facial features or clothing colour.

table

Join us

Join our group if you are interested in exploring the exciting possibilities of drone swarm technology and its potential applications in areas such as disaster response, environmental monitoring, and smart cities. You may can also find information on our Computer Science web pages about how to join us for academic visits.

For PhD or other research opportunities, please direct enquiries to Nadia Kanwal (n.kanwal@keele.ac.uk).

Publications

Yousuf, M. J., Lee, B., Asghar, M. N., Ansari, M. S., & Kanwal, N. (2024). Unlocking trust: Advancing activity recognition in video imagery. IEEE Access, 12, 176799–176817. doi: 10.1109/ACCESS.2024.3503284.
Tahir, M., Qiao, Y., Kanwal, N., Lee, B., & Asghar, M. N. (2023). Real-time event-driven road traffic monitoring system using CCTV video analytics. IEEE Access, 11, 139097–139111. doi: 10.1109/ACCESS.2023.3340144.
ahir, M., Qiao, Y., Kanwal, N., Lee, B., & Asghar, M. N. (2023). Privacy-preserved video summarisation of road traffic events for IoT smart cities. Cryptography, 7(1), org/10.3390/cryptography7010007
Yousuf, M. J., Kanwal, N., & Ansari, M. S. (2022). Deep learning-based human detection in privacy-preserved surveillance videos. 35th International BCS Human-Computer Interaction Conference. DOI: 10.14236/ewic/HCI2022.33
Aribilola, I., Asghar, M. N., Kanwal, N., Fleury, M., & Lee, B. (2023). SecureCam: Selective detection and encryption-enabled application for dynamic camera surveillance videos. IEEE Transactions on Consumer Electronics, 69(2), 156–169., doi: 10.1109/TCE.2022.3228679.
Zaidi, S. S. A., Ansari, M. S., Aslam, A., Kanwal, N., Asghar, M., & Lee, B. (2022). A survey of modern deep learning-based object detection models. Digital Signal Processing, 126, 103514. org/10.1016/j.dsp.2022.103514
Amna, S., Asghar, M. N., Kanwal, N., Ansari, M. S., Fleury, M., Lee, B., Herbst, M., & Qiao, Y. (2020). MuLViS: Multi-level encryption-based security system for surveillance videos. IEEE Access, 8, 177131–177155. doi: 10.1109/ACCESS.2020.3024926.
Kanwal, N., Asghar, M. N., Ansari, M. S., Fleury, M., Lee, B., Herbst, M., & Qiao, Y. (2020). Preserving chain-of-evidence in surveillance videos for authentication and trust-enabled sharing. IEEE Access, 8, 153413–153424. doi: 10.1109/ACCESS.2020.3016211.