Artificial intelligence inspired framework for preventing sexual violence at public toilets of educational institutions with the improvisation of gender recognition from gait sequences.

Sexual violence against female students in public toilets is one of the major concerns for educational institutes. With the advent of smart technologies and smart campuses, educational institutes are making the best of their efforts to prevent such events from occurring. Existing approaches have shortcomings such as a low recognition time, locating errors, and alarming latency. In this study, we aim to propose an Artificial Intelligence (AI) inspired framework to prevent sexual violence in public toilets of educational institutions with the improvisation of gender recognition from gait sequences by collecting the data from IoT devices installed at the public toilets. Gait recognition provides the ability to observe an individual unremarkably, without any direct cooperation from the human. In this study, we have used the fusion of Speeded-Up Robust Features (SURF) and Convolutional Neural Networks (CNN) for feature extraction and Support Vector Machine (SVM) as a classifier. The performance of the proposed framework is compared with other state-of-the-art approaches. The results indicate that the proposed framework outperforms all other considered approaches, and attained a higher accuracy (95.7%). [ABSTRACT FROM AUTHOR]