Driver Emotions Recognition Based on Improved Faster R-CNN and Neural Architectural Search Network

Abstract

It is critical for intelligent vehicles to be capable of monitoring the health and well-being of the drivers they transport on a continuous basis. This is especially true in the case of autonomous vehicles. To address the issue, an automatic system is developed for driver’s real emotion recognizer (DRER) using deep learning. The emotional values of drivers in indoor vehicles are symmetrically mapped to image design in order to investigate the characteristics of abstract expressions, expression design principles, and an experimental evaluation is conducted based on existing research on the design of driver facial expressions for intelligent products. By substituting a custom-created CNN features learning block with the base 11 layers CNN model in this paper for the development of an improved faster R-CNN face detector that detects the driver’s face at a high frame per second (FPS). Transfer learning is performed in the NasNet large CNN model in order to recognize the driver’s various emotions. Additionally, a custom driver emotion recognition image dataset is being developed as part of this research task. The proposed model, which is a combination of an improved faster R-CNN and transfer learning in NasNet-Large CNN architecture for DER based on facial images, enables greater accuracy than previously possible for DER based on facial images. The proposed model outperforms some recently updated state-of-the-art techniques in terms of accuracy. The proposed model achieved the following accuracy on various benchmark datasets: JAFFE 98.48%, CK+ 99.73%, FER-2013 99.95%, AffectNet 95.28%, and 99.15% on a custom-developed dataset.

Research on optimization algorithm of abnormal driving behavior detection based on PFLD

“…5. The world coordinate system to camera coordinate system transformation is shown in formula (7), the camera coordinate system to pixel coordinate system transformation and the relation between world and pixel coordinates are shown in formulas ( 8) and ( 9), and formula ( 9) can be solved iteratively by a combination of least squares and direct linear transformation, where the least squares objective function is shown in formula (10), without " ^ " for measured values and with " ^ " for predicted values.…”

“…Zaman Khalid et al proposed a DRER based on the NasNet-Large CNN architecture. By combining the improved R-CNN and transfer learning, the accuracy of DER based on facial images could reach 99% [10]. Although these recognition methods have achieved outstanding results in the laboratory, it still has problems such as large computational load, speed mismatch, limited application range and low efficiency when the model is deployed to the mobile terminal.…”

Deep Transfer Learning Enabled Intelligent Object Detection for Crowd Density Analysis on Video Surveillance Systems

“…Neural architecture search (NAS) is the most important search technique that the authors deployed in this network. Child network is shown to accomplish some accuracy on a validation set; in other words, it is used for convergence [25]. The subsequent accuracy value is utilized to upgrade the controller that consecutively generates better architecture over time.…”

SS-MFAR : Semi-supervised Multi-task Facial Affect Recognition

“…Automatic affect recognition is currently an active area of research and has applications in many areas such as education, gaming, software development, auto motives [1,2], medical care, etc. Many works have dealt with Valence-arousal estimation [3,4,5,6,7], action unit detection [8,9,10], expression classification [] tasks individually.…”