Date of Submission
Master of Science in Computer Science (MSCS)
Dr. Selena He
Dr. Dan Lo
Dr. Chih Cheng Hung
Dr. Mingon Kang
Dr. Selena He
Evaluation of traditional classroom has led to electronic classroom i.e. e-learning. Growth of traditional classroom doesn’t stop at e-learning or distance learning. Next step to electronic classroom is a smart classroom. Most popular features of electronic classroom is capturing video/photos of lecture content and extracting handwriting for note-taking. Numerous techniques have been implemented in order to extract handwriting from video/photo of the lecture but still the deficiency of few techniques can be resolved, and which can turn electronic classroom into smart classroom.
In this thesis, we present a real-time IoT system to convert handwritten text into editable format by implementing hand gesture recognition (HGR) with Raspberry Pi and camera. Hand Gesture Recognition (HGR) is built using edge detection algorithm and HGR is used in this system to reduce computational complexity of previous systems i.e. removal of redundant images and lecture’s body from image, recollecting text from previous images to fill area from where lecture’s body has been removed. Raspberry Pi is used to retrieve, perceive HGR and to build a smart classroom based on IoT. Handwritten images are converted into editable format by using OpenCV and machine learning algorithms. In text conversion, recognition of uppercase and lowercase alphabets, numbers, special characters, mathematical symbols, equations, graphs and figures are included with recognition of word, lines, blocks, and paragraphs. With the help of Raspberry Pi and IoT, the editable format of lecture notes is given to students via desktop application which helps students to edit notes and images according to their necessity.
patel, Nidhi, "An IoT System for Converting Handwritten Text to Editable Format via Gesture Recognition" (2018). Master of Science in Computer Science Theses. 13.