International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 12 | Dec 2025
p-ISSN: 2395-0072
www.irjet.net
Emotion Tracker: Real-time Facial Emotion Detection with Open CV and Deep Face Prof.Ashwini Sangam 1, Prajwal2 1Professor, Master of Computer Application, VTU’s CPGS, Kalaburagi, Karnataka, India 2Student, Master of Computer Application, VTU’s CPGS, Kalaburagi, Karnataka, India
----------------------------------------------------------------------------***----------------------------------------------------------------------------algorithms are either computationally intensive, reliant on cloud storage, or designed to work with still images only. Facial Emotion Detection platform using open CV Therefore, a low-cost and rapid option for online detection (Computer Vision) and Deep Face (AI-based Emotion of facial expressions is needed that can generate accurate Classification). This platform will take a live webcam video and time-sensitive results from a video feed, as well as feed and analyze it to detect the face of the subject(s) and allow for an easy and intuitive viewing experience. To classify their emotional state (Happy, Sad, Angry, Fearful, address this issue, the Emotion Tracker Project utilizes Surprised, Disgusted, or Neutral). The classification will both Open CV and Deep Face. occur in real time and once completed, the classification can then be logged to the database for future analysis. 3. OBJECTIVES Approximately 90% accurate, reasonably priced, easy to use and capable of operating in "Offline Mode", this platform Emotion Tracker aims to build an Emotion detection can be beneficial for a variety of industries (Healthcare, system that captures facial expressions in real-time using Education, Customer Service, and Security) in numerous Open CV and Deep Face. It will capture human face images ways. from live-streaming video data, accurately detect and classify 6 Different Emotions (using facial recognition Keywords: Facial Emotion Recognition, Real-Time technology) on real-time and capture emotion information Detection, Deep Learning, Open CV, Deep Face, for later analysis. The Emotion Tracker will also be Computer Vision, HCI. designed to be extensible and modular to allow for future adaptations (including integration into other 1. INTRODUCTION applications). Traffic congestion is an expanding challenge in cities, often 4. RESEARCH METHODOLOGY resulting in extensive delays and poor vehicle movement. Fixed-time traffic signals, the default for controlling car flow, cannot respond to fluctuating conditions making the The project uses a systematic process that consists of traffic signal ineffective at peak hours or unexpected requirement analysis, system design, implementation, surges. Emerging technologies, specifically in computer testing, and deployment. The webcam captures the video vision and machine learning, make it possible to remotely stream live; Open CV detects faces in the video, and Deep and in real-time analyse traffic flow without on-site Face is used for emotion classification. The system will be inspection using video based detection, tracking, and tested in various conditions to confirm that it operates in density estimation. The intelligent signal control system real-time with high levels of accuracy and dependability. proposed here employs intelligent signal control techniques using computer vision and machine learning to 5. REVIEW OF LITERATURE develop simulation models that can assess car flow and dynamically apply signal timing using a max-pressure At first, early systems for identifying facial expressions algorithm. The ultimate goal is to reduce wait times, used hand-constructed characteristics and conventional reduce congestion and increase the efficiency of traffic algorithms. However, these systems did not offer very management. reliable performance in a real-world environment. With the introduction of deep learning and Convolutional 2. PROBLEM STATEMENT Neural Networks (CNNs) into the space, the effectiveness of identifying emotions increased significantly. Pre-trained Detecting facial expressions in real time is difficult models like VGG-Face and tools such as Deep Face allow because of the many factors involved (e.g. the way for much quicker and more precise construction of someone is feeling, the type of lighting they are in, their emotion detection systems. Nevertheless, despite the angle of view, and whether they are hiding their face). availability of various developed solutions today, many are Most traditional techniques are not very accurate or still expensive to run, require usage of the cloud, and robust, while the vast majority of modern deep learning therefore cannot be used for real-time applications. That is
Abstract-The "Emotion Tracker" Project is a real Time
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 204