Indonesian Journal of Electrical Engineering and Computer Science Vol. 30, No. 1, April 2023, pp. 598~605 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v30.i1.pp598-605  598 Journal homepage: http://ijeecs.iaescore.com Analysis of facial emotion recognition rate for real-time application using NVIDIA Jetson Nano in deep learning models Usen Dudekula, Purnachand N School of Electronics Engineering, VIT-AP University, Amaravati, Andhra Pradesh, India. Article Info ABSTRACT Article history: Received May 14, 2022 Revised Dec 15, 2022 Accepted Dec 20, 2022 Detecting facial emotion expression is a classic research problem in image processing. Face expression detection can be used to help human users monitor their stress levels. Perceiving an individual's failure to communicate specific looks might help analyze early psychological disorders. several issues like lighting changes, rotations, occlusions, and accessories persist. These are not simply traditional image processing issues, yet additionally, action units that make gathering activity of facial acknowledgment troublesome look information, and order of the demeanor. In this study, we use Xception taking into account Xception and convolution neural network (CNN), which is easy to focus on incredible parts like the face, and visual geometric group (VGG- 19) used to extract the facial feature using the OpenCV framework classifying the image into any of the basic facial emotions. NVIDIA Jetson Nano has a high video handling outline rate. Accomplishing preferable precision over the recently evolved models on software. The average accuracies for standard data set CK+,” on NVIDIA Jetson Nano, the accuracy rate is 97.1% in the Xception model in the convolutional neural network, 98.4% in VGG-19, and real-time environment accuracy using OpenCV, accuracy rate is 95.6%. Keywords: Deep learning NVIDIA Jetson Nano Transfer learning Visual geometry group Xception Architecture This is an open access article under the CC BY-SA license. Corresponding Author: Purnachand N School of Electronics Engineering, VIT-AP University G-30, Inavolu, Beside AP Secretariat Amaravati, Andhra Pradesh 522237, India Email: chanduinece@gmail.com 1. INTRODUCTION Acknowledgment of facial feelings is the strategy for recognizing human sentiments through looks. The human mind intuitively perceives feelings, and programming has now been made that can likewise perceive emotions we use Xception considering Xception and convolution neural network (CNN), which is easy to zero in on inconceivable parts like the face. Time, this technology is becoming more specific, and will finally be able to read emotions as our brains do. Using facial expressions and vocal tones, people usually interpret the profound emotional conditions of others, for example, excitement, sorrow, and rage. According to a survey [1]. The rapid development of artificial intelligence techniques for automatic facial expression processing (FER), including human-PC association (HCI), computer-generated reality (VR), increased reality (AR), the high-level driver helps frameworks (ADASs), and diversion. Consequently, various sensors, like electromyography (EMG), electrocardiogram (ECG), and camera is the most encouraging kind of sensor since it offers the most nitty gritty hints, for facial feeling acknowledgment (FER) inputs [2]–[7]. Due to a variety of challenges faced in the identification and recognition of lighting and accessories, partial occlusions, head deviation of the facial regions, and a low identification rate, FER remains difficult. Deep learning methodologies may offer a suboptimal solution to these problems [8], [9]. GPUs are also utilized in other applications that benefit from their parallel nature, including machine learning and deep learning applications.