International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024
p-ISSN: 2395-0072
www.irjet.net
Facial Expression Recognition in Advanced Driving Assistance System Dhruv Nadkar1, Aditya Avhad2, Bhushan Gajare3 1Professor V.R. Jaiswal, Pune Institute of Computer Technology, Pune, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In the automobile domain, advanced
drowsiness detection, Electronic Stability Control (ESC), Forward Collision Warnings (FCW), Lane Departure Warning System (LDWS), and Traffic Sign Recognition (TSR) are examples of ADAS technology. The majority of ADASs are electronic systems that adapt and improve vehicle safety and driving quality. By correcting for human mistakes, they have been shown to minimize road deaths. Proposed System uses CNN and deep learning algorithms to detect and predict facial emotion of drivers by keeping track of images and through videos.
driver-assistance systems (ADASs) are employed to improve safety, however existing ADASs do not take into consideration drivers' conditions, such as whether they are emotionally suited to drive. In the automotive industry, advanced driver-assistance systems (ADASs) are employed to improve safety, however existing ADASs do not consider drivers' circumstances, such as whether they are emotionally fit to drive. Many road accidents and unanticipated situations are caused by driver inattention, which is one of the key characteristics and reasons.Face expression recognition is a relatively new image processing technique that is becoming increasingly important in applications such as driver warning systems. Even when given a noisy input or incorrect data, current algorithms can recognise facial expressions, but they lack accuracy. It's also useless when it comes to dealing with uncontrollable emotions and recognition. Based on Deep Neural Networks and Convolutional Neural Networks, the proposed technique provides a driver warning system that efficiently identifies face expressions (CNN).
2. LITERATURE SURVEY Face recognition has gotten a lot of attention in recent years as one of the most successful uses of image analysis and comprehension. Automatic FER approaches have been extensively researched for many years, and because the use of the most discriminative options is the most important factor determining a FER method's effectiveness, they'll be divided into two categories: those using handsewn options and those using options generated by a deep learning network. One of the most prominent uses of FER is in the Advanced Driving Assistance System.
Key Words: Convolutional Neural Networks (CNNs), Deep Learning (DL), Driver Warning System, Facial Emotion Recognition (FER), Advance Driving Assistance System (ADAS).
The driver's gaze was fixed on the specially constructed drowsy driver detection system, which was utilized to assess weariness. Techniques for detecting drowsiness are divided into two groups based on the criteria employed for detection: intrusive and non-invasive methods of detection. The distinction is made depending on whether or not an instrument is attached to and paired with the driver. An instrument is well-connected to the driver in the invasive technique, and the value of that instrument is examined and recorded.
1. INTRODUCTION Excessive driver distractions, alcohol consumption, and speeding beyond safe limits are widely recognized as major causes of road accidents and mishaps. According to statistics from respective departments, it is observed, another very crucial factor contributing to road accidents is the fatigue condition that a driver experiences while driving. Drivers experiencing mental fatigue also suffer from excessive sleepiness and loss of consciousness after regular intervals. Drivers driving for more than 8 hours a day, undergoing immense physical activities and lack of sleep usually suffer mental fatigue.
In their paper, S. Suchitra, S. Sathya Priya, R. J. Poovaraghan, B. Pavithra, and J. Mercy Faustina [1] propose a Local Octal Pattern-Convolutional Neural Network (LOP-CNN) approach to develop a more efficient Driver Warning System using Deep Learning. The CNNbased feature extraction reduces the semantic gap and enhances overall performance by utilizing Facial Expression Recognition. In their paper, Mira Jeong and Byoung Chul Ko [2], explain that ADAS integrates psychological models, sensors to capture physiological data, human emotion categorization algorithms, and human-car interaction algorithms. Researchers are focusing on issues such as using subtle psychological and physiological indicators (e.g., eye closure) to enhance the detection of dangerous situations like distraction and
In recent years, a rising amount of automation has penetrated the automobile industry. In terms of manual driving, this automation has opened up new options. On-board Advanced Driver-Assistance Systems (ADASs), which are used in automobiles, trucks, and other vehicles, provide exceptional opportunities for increasing the quality of driving, safety, and security for both drivers and passengers. Adaptive Cruise Control (ACC), Anti-lock Braking System (ABS), automotive night vision,
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 562