Skripsi
ANALISIS PENGENALAN POLA HURUF DAN ANGKA RUANGAN MENGGUNAKAN METODE WEIGHTLESS NEURAL NETWORK
Pattern recognition is an essential aspect of artificial intelligence with various applications, including face, voice, and character recognition. In the healthcare field, this technology can support navigation systems for medicine delivery robots, reducing direct physical contact between medical staff and patients with infectious diseases. This research focuses on applying the Weightless Neural Network (WNN) method for character and number recognition using Raspberry Pi as the detector and binary data converter, and Nucleo F401RE microcontroller as the data processor. The objectives are to identify the implementation of WNN on hardware devices, analyze the effectiveness of Nucleo F401RE in binary data processing, and evaluate WNN’s performance in pattern recognition. The results show that the system successfully detected text using Tesseract OCR, which was processed into a 75 × 45 binary array and stored in 16 datasets. WNN achieved recognition accuracy of 95.83%. The Nucleo F401RE also demonstrated excellent processing speed, ranging from 0.01 to 0.02 seconds, supporting real-time implementation. Therefore, this system has strong potential to be applied in medicine delivery robots to enhance healthcare efficiency.
| Inventory Code | Barcode | Call Number | Location | Status |
|---|---|---|---|---|
| 2507006193 | T185479 | T1854792025 | Central Library (Reference) | Available but not for loan - Not for Loan |