Stuart J. Gibson

TL;DR: In this paper, a convolutional neural network is trained to automatically identify substances according to their Raman spectrum without the need for preprocessing, and superior classification performance is demonstrated compared with other frequently used machine learning algorithms including the popular support vector machine method.

TL;DR: A unified solution for the identification of chemical species in which a convolutional neural network is trained to automatically identify substances according to their Raman spectrum without the need for preprocessing is described.

TL;DR: This paper introduces DeepCAPTCHA, a new and secure CAPTCHA scheme based on adversarial examples, an inherit limitation of the current DL networks, and implements a proof of concept system, which shows that the scheme offers high security and good usability compared with the best previously existing CAPTCHAs.

TL;DR: This work presents a statistically rigorous approach to the aging of digitised images of the human face based on the calculation of optimised aging trajectories in a model space and aged images can be obtained through a fast, semi-automatic procedure.

TL;DR: Preliminary examples of composites generated with the facial composite system are presented which demonstrate the potential superiority of the evolutionary approach to composite generation.