Gender Classification Hand Recognition System Using Key-Point Detection with Deep Neural Network
Rajbir Singh Girn, Philippe Garrine Dimson Lim, Jocelyn F. Villaverde · 2021 IEEE 13th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) · 2021
Classification Systems evolve every day with new features, new engines, and updated hardware and software. It indeed depends on what type of system is being assembled and for what purpose. These classification systems will require some biometric-based mechanic to know what reference to identify a particular object or individual. These types of classification systems can be helpful within the public areas, like determining whether that person of interest in the video is a man or a woman. This can help authorities much more outstanding and can assist with their investigations. Our main objective is to classify, label, and compare the recommended biometric used in the gender classification system. This includes knowing the results like the classification type, accuracy, response time, type of biometric, and method used, like in this case, by using the key-point detection and the DNN. Overall, the researchers concluded that the gender classification system using hand recognition has a high accuracy rate with an overall total of 88.2353%. Including a response rate of around 3-4 seconds from a total of 17 test trials.