Print Email Facebook Twitter Towards a low-cost air-written character recognition system Title Towards a low-cost air-written character recognition system: Designing an embedded machine learning system to recognise the first 10 letters of the Latin alphabet Author Pronk, Paco (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Wang, Q. (mentor) Yang, M. (graduation committee) Zhu, R. (graduation committee) Degree granting institution Delft University of Technology Programme Computer Engineering Date 2023-06-25 Abstract This study introduces a novel system that leverages three photodiodes and ambient light to identify air-written characters on a resource-constrained device. Through experimentation, suitable methods of data preprocessing, machine learning and model compression were selected to recognize the first 10 characters of the Latin alphabet. The final system was able to recognize these characters with a between-participant accuracy of 50.80% and a within-participant accuracy of 67.82%. Subject TinyMLEmbedded Deep LearningTouch-Free InteractionGesture RecognitionAir WritingVisible LightTensorFlow Lite for Microcontrollersneural networkconvolutional neural networkArduino Nano 33 BLE To reference this document use: http://resolver.tudelft.nl/uuid:0dbeb7e6-89bd-41e5-bf1d-f99363702551 Part of collection Student theses Document type bachelor thesis Rights © 2023 Paco Pronk Files PDF Final_BEP_report_Paco_Pronk.pdf 7.2 MB Close viewer /islandora/object/uuid:0dbeb7e6-89bd-41e5-bf1d-f99363702551/datastream/OBJ/view