Skripsi
IMPLEMENTASI PENGENALAN WAJAH PADA MIKROKONTROLLER DENGAN METODE WEIGHTLESS NEURAL NETWORKS DENGAN IDENTIFIKASI MATA DAN MULUT
In this study, the only parts of the face used were the eyes and mouth. Apart from that, data in facial recognition is also limited, namely a distance of 40-50cm from the camera. This facial recognition is done on a microcontroller. WNN was chosen because of its ability to reduce memory usage, while to measure similarity using the immediate scan method. Immediate scan was chosen because of its simplicity and speed in measuring similarity. Testing and implementation uses servo for face tracking. This research uses a mini computer to take images to convert them into binary images and send them to the microcontroller so that the microcontroller can process the image and recognize it. In this study, the dataset used is a primary dataset containing the author's face, 10 eye data and 10 mouth data. The 20 data are embedded in the microcontroller to be used as a dataset. Based on the tests that have been carried out, of the 10 images that fit the dataset, all of them were recognized with percentage results between 81.31% - 91.88% for the eyes and 82.18% - 93.90% for the mouth. Meanwhile, testing on 20 images outside the dataset produced 42.70% - 56.96% for the eyes and 53.24% - 72.21% for the mouth, so they were considered unrecognizable.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407000813 | T139308 | T1393082024 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available