• N Chakridhar
• Esther Keerthi Suravarapu
• Dandu Damarudh Varma

The advancement of computer technology has revolutionized the way people interact with digital devices. However, individuals with physical disabilities face significant barriers to accessing these technologies due to reliance on conventional input devices like keyboards and mice. In this study, we propose a deep learning-based system that enables hands-free computer interaction through eye-tracking technology, allowing cursor control solely via eye movements. Utilizing a standard laptop camera, this solution eliminates the need for additional hardware, making it both cost-effective and accessible. Our system combines Haar Cascade Classifiers and Convolutional Neural Networks (CNN) to detect and interpret eye positions and movements, translating these into precise cursor actions and click commands using OpenCV. Through real-time eye aspect ratio (EAR) measurements, the system accurately recognizes intentional eye gestures, such as blinks and gaze shifts, that control cursor movement and simulate mouse clicks. This approach allows individuals with limited mobility or neuromuscular conditions to navigate and interact with computers independently, fostering digital inclusion and enhancing quality of life. The proposed model is highly adaptable and could be further refined to support additional functionalities, such as text input and remote device control. By broadening accessibility to essential digital tools, this research marks a significant step forward in Human-Computer Interaction (HCI) and offers a versatile platform for developing advanced, non-contact control interfaces tailored to user’s needs.

Haar cascade, CNN, EAR (Eye Aspect Ratio)

"Eyeball Cursor Movement using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 11, page no.c555-c570, November-2024, Available :http://www.jetir.org/papers/JETIR2411274.pdf

"Eyeball Cursor Movement using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 11, page no. ppc555-c570, November-2024, Available at : http://www.jetir.org/papers/JETIR2411274.pdf