Raspberry Based Hand Gesture Recognition Using Haar Cascade and Local Binary Pattern Histogram

Abstract

Many companies and even public institutions for civil servants currently use photo-taking for the attendance. However, this strategy is still considered ineffective since the employees still can hack the attendance by making their own photos and put them in their desks. Therefore, an alternative that can complement the current face detection method is highly needed so that the employee’s attendance can be directly monitored. One of the methods that can be used to detect the attendance is hand gesture detection. This research aims to detect hand gestures made by the employees to ensure whether they really come to work or not. This research make  the chance for manipulation using photo or fake GPS is quite small. For the purpose of hand gesture recognition, this study utilized Local Binary Pattern Histogram algorithm. The hand gesture image was first taken using a raspberry pi camera and then processed by the device to examine whether it matches the registered ID or not. The results showed that ID recognition by using hand gestures is detectable. The number recognition in hand gestures includes numbers 1 to 10. The test results showed that for 5 trials, the average time required for reading hand gestures using a laptop was 9.2 seconds, while that of using raspberry was 14.2 seconds. The results of this research show that the system has not been able to distinguish which hand is read first, so numbers that have the same number are considered the same, such as 81 and 18. So, the motion reading using a raspberry takes longer than that of using a laptop because the laptop's performance is higher than that of a raspberry and system cannot distinguish between numbers consisting of the same number.