Design and Development of Student Attendance Application Using Viola-Jones Methodand Haar Cascade Algorithm at UIM

Abstract

The recognition of human facial images is one of the key technologies that continues to be developed. In the field of Computer Vision, it finds its applications in biometric recognition systems, search and indexing in digital video databases, restricted area access control systems, video conferences, human-computer interaction, and more. The Viola-Jones method is an object detection method known for its high accuracy of approximately 93.7%, and it is 15  times  faster  than  the  Rowley  Baluja-Kanade  detector  and  around  600  times  faster  than  the  Schneiderman-Kanade  detector.  The  Haar  Cascade  Classifier  algorithm  is  one  of  the  algorithms  used  to  detect  a  face.  The Cascade  Classifier  is  utilized  in  attendance  recording  with  real-time  face  recognition,  enabling  the  real-time recording of students.To enhance the abstract, it could be mentioned that to further improve the effectiveness and accuracy  of face  recognition  systems,  researchershave  explored  combining  the  Viola-Jones  method  with  other techniques. For instance, the Haar Cascade Classifier algorithm can be combined with the Local Binary Pattern (LBP)  algorithm,  which  preserves  important  information  in  images  and  performs  well  under  low  lighting conditions.  By  integrating  these  techniques,  face  recognition  systems  can  achieve  better  results  in  classifying human heads, which are often the primary objects of interest in face recognition. Additionally, in conducting photo capture experiments using a camera, it is important to employ grayscale coloration, which retains the grayscale information in the images. Furthermore, capturing six photos during the process ensures more effective results and facilitates the assignment of unique IDs for easier data differentiation.