Study of the Precision and Feasibility of Facial Recognition using OpenCV with Java for a System of Assistance Control

Computer vision is a computational discipline which attempts to emulate human vision and has been done presently in current systems and applications. Facial recognition is a biometric method that allows identifying different people by means of unique features that the human eye is simply unable to perceive. Currently, most systems and applications require greater security, the reliability of use and protection of data, which is why we opt for the use of technologies based on facial recognition as the case of Google and Facebook, to name a few. This face detection technology is mainly based on visible images and despite being a discipline with more than 30 years of study, still have certain problems at the implementation time, such as the light under which the images are captured, the posture, face expressions and the quality with which the device captures the image, among others. However, there is a wide variety of it tools that allow making improvements in the image and the detection of faces in an easier way; among the most used are MatLab and Phyton that employ the use of open source libraries as OpenCV in the case of this study, has been opted for the use of Java in conjunction with OpenCV, Because unlike MatLab, it does not represent economic cost and facilitates the integration of computer vision algorithms with an assistance control system.