Developing testing frameworks for AI cameras

It is possible for inexpensive cameras to include AI based features such as face recognition. However a test framework for such cameras is required that will allow comparison of accuracy under differing conditions. This will then lead to the improvement of training data and algorithms.
A simple test framework has been developed and partly evaluated by testing multiple head/face accessories under different lighting conditions using 6 participants with over 300 pictures using a Husky Lens. It was found that it can detect faces at a reasonable level of accuracy during ‘middle of the day’ lighting conditions with or without head accessories. However it delivers significantly lower detection rate with accessories that cover parts of the face and under green light.
There is still a need to further investigate this area of study with a higher number of testing participants compared to the 6 participants that have been used for this research. It is anticipated that better testing frameworks will lead to better algorithms, training data and specifications for users.

This version of the paper has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use (https://www.springernature.com/gp/open-research/policies/accepted-ma...), but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/10.1007/978-3-031-21441-7_28

Cite this paper:
Herdzik, A., James-Reynolds, C. (2022). Developing Testing Frameworks for AI Cameras. In: Bramer, M., Stahl, F. (eds) Artificial Intelligence XXXIX. SGAI-AI 2022. Lecture Notes in Computer Science(), vol 13652. Springer, Cham. https://doi.org/10.1007/978-3-031-21441-7_28