• Nanda B S
• Ankitha H R
• Sahana H S
• Arun Kumar V
• Shradha holagi

The article discusses the development of an air gesture keyboard using a machine learning approach to assist visually impaired individuals. The keyboard utilizes a remote-based system that translates hand gestures into text through an accelerometer, Arduino, and switches. It eliminates the need for physical keyboards and supports multi-linguistic functionality. The article begins with an introduction highlighting the significance of keyboards in data entry and the prevalence of touch and gesture keyboards. It emphasizes the need for alternative input methods for visually impaired individuals and the limitations of existing technologies. The proposed remote-based gesture keyboard enables text input by capturing hand motions in the air, with the remote acting as a pen and a text editor as a notebook. A literature survey is conducted to explore related research on hand gesture recognition using different techniques such as digital cameras, EMG signals, and sensor-based approaches. The technical requirements for the air gesture keyboard system are then discussed, including the accelerometer MPU 6050, Arduino UNO, button switch, and Bluetooth device. The system design is presented, illustrating the circuit diagram and the creation of gesture recognition software using the Scikit-learn library. The article also discusses the various stages of the machine learning algorithm, including dataset creation, data pre-processing, training set formation, and algorithm selection. Classification algorithms, specifically Support Vector Machines (SVM), are evaluated for gesture recognition. Overall, the proposed air gesture keyboard offers a novel and efficient input method for visually impaired individuals, allowing them to enter text without physical keyboards. The system's machine learning capabilities enhance accuracy and versatility, making it a promising assistive technology for visually impaired individuals.

machine learning, gesture, Microcontroller, visually impaired

"AIR GESTURE KEYBOARD FOR VISUALLY IMPAIRED PEOPLE USING MACHINE LEARNING APPROACH", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 6, page no.j355-j363, June-2023, Available :http://www.jetir.org/papers/JETIR2306949.pdf

"AIR GESTURE KEYBOARD FOR VISUALLY IMPAIRED PEOPLE USING MACHINE LEARNING APPROACH", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 6, page no. ppj355-j363, June-2023, Available at : http://www.jetir.org/papers/JETIR2306949.pdf