• Mrs Susmitha M N
• Veena M V
• Vidyashree T C
• Varshini Yenduri
• Shalini K Naik

Human falls could be highly dangerous especially when the fallen people lay unattended or unobserved for a long time. Depending on the impact of the fall, the casualties may also increase. Prolonged blood loss can also lead to early death or major health problems. Thus, it is necessary to take measures to help the victims, who unfortunately are mostly old or disabled. Even after showing good accuracy rates, wearable sensor-based fall detection systems are not used by many potential users due to a variety of personal and physical reasons. Thus, detection of falls using cameras is a very reliable solution to address this problem This paper proposes an efficient fall detection system based on camera vision using convolutional neural networks (CNN). Convolutional Neural Networks have proven to be an efficient means of classification in the field of image processing. Here we propose an efficient fall detection design using Inception model CNN by applying transfer learning and aim to achieve state-of-the-art accuracy in multiple datasets such as URFD (UR Fall Detection Dataset) and FDD(Fall Detection Dataset)

Detecting elderly behavior based on deep learning for healthcare, detecting elderly, detecting elderly behavior on deep learning, detecting elderly behavior on

"DETECTING ELDERLY BEHAVIOR BASED ON DEEP LEARNING FOR HEALTH CARE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 5, page no.a638-a641, May-2023, Available :http://www.jetir.org/papers/JETIR2305090.pdf

"DETECTING ELDERLY BEHAVIOR BASED ON DEEP LEARNING FOR HEALTH CARE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 5, page no. ppa638-a641, May-2023, Available at : http://www.jetir.org/papers/JETIR2305090.pdf