Experimental study on visual detection for fatigue of fixed-position staff

On Fatigue Detection for Air Traffic Controllers Based on Fuzzy Fusion of Multiple Features

Fatigue detection for air traffic controllers is an important yet challenging problem in aviation safety research. Most of the existing methods for this problem are based on facial features. In this paper, we propose an ensemble learning model that combines both facial features and voice features and design a fatigue detection method through multifeature fusion, referred to as Facial and Voice Stacking (FV-Stacking). Specifically, for facial features, we first use OpenCV and Dlib libraries to extract mouth and eye areas and then employ a combination of M-Convolutional Neural Network (M-CNN) and E-Convolutional Neural Network (E-CNN) to determine the state of mouth and eye closure based on five features, i.e., blinking times, average blinking time, average blinking interval, Percentage of Eyelid Closure over the Pupil over Time (PERCLOS), and Frequency of Open Mouth (FOM). For voice features, we extract the Mel-Frequency Cepstral Coefficients (MFCC) features of speech. Such facial features and voice features are fused through a carefully designed stacking model for fatigue detection. Real-life experiments are conducted on 14 air traffic controllers in Southwest Air Traffic Management Bureau of Civil Aviation of China. The results show that the proposed FV-Stacking method achieves a detection accuracy of 97%, while the best accuracy achieved by a single model is 92% and the best accuracy achieved by the state-of-the-art detection methods is 88%.

Visual responses of patients with generalized anxiety disorder who cycling in the virtual sportscapes with different tree cover densities

The tree density of virtual sportscape is the main factor that determines the benefits that generalized anxiety disorder (GAD) patients can obtain when they exercise with virtual environment. By using pupil size, fixation count and time as metrics, this research aimed to clarify the relationship between tree cover density and stress in the virtual environment. Ninety GAD patients were randomly grouped into the 36-60% tree density (high tree density, HTDS), 20-35% tree density (medium tree density, MTDS), or control groups (n = 30). Researchers used eye-tracking technology to analyze fixation time, fixation count and changes in pupil size to evaluate the stress changes of participants after 20 min of aerobic exercise in a virtual environment. The results showed that pupil size expanded in GAD patients after exercising in the virtual environment. Furthermore, GAD patient cycling in the MTDS group can show smaller pupil size than those in HTDS. Those results suggest that GAD patient cycling 20 min in the MTDS group can perceived lower stress. The results of eye tracking analysis showed that GAD patients spend more time and counts observing tree elements in HTDS and MTDS sportscapes. Specifically, they spent more 48% and 27% time on tree and green plants in the HTDS condition and MTDS condition, respectively, than in non-natural sportsscapes. Although 36-60% tree density of virtual natural sportscape can get more visual attention from GAD patients, 20-35% tree density of virtual natural sportscape is more capable of reducing their stress.

A High-Efficiency Fatigued Speech Feature Selection Method for Air Traffic Controllers Based on Improved Compressed Sensing

Air traffic controller fatigue has recently received considerable attention from researchers because it is one of the main causes of air traffic incidents. Numerous research studies have been conducted to extract speech features related to fatigue, and their practical utilization has achieved some positive detection results. However, there are still challenges associated with the applied speech features usually being of high dimension, which leads to computational complexity and inefficient fatigue detection. This situation makes it meaningful to reduce the dimensionality and select only a few efficient features. This paper addresses these problems by proposing a high-efficiency fatigued speech selection method based on improved compressed sensing. For adapting a method to the specific field of fatigued speech, we propose an improved compressed sensing construction algorithm to decrease the reconstruction error and achieve superior sparse coding. The proposed feature selection method is then applied to optimize the high-dimension fatigued speech features based on the fractal dimension. Finally, a support vector machine classifier is applied to a series of comparative experiments using the Civil Aviation Administration of China radiotelephony corpus to demonstrate that the proposed method provides a significant improvement in the precision of fatigue detection compared with current state-of-the-art approaches.

Improving Eye-Computer Interaction Interface Design: Ergonomic Investigations of the Optimum Target Size and Gaze-triggering Dwell Time

Interactive feedback of interface elements and low level of spatial accuracy are two main key points for the interaction research in the Eye-computer interaction system. This study tried to solve these two problems from the perspective of human–computer interactions and ergonomics. Two experiments were conducted to explore the optimum target size and gaze-triggering dwell time of the eye–computer interaction (ECI) system. Experimental Series 1 was used as the pre-experiment to identify the size that has a greater task completion rate. Experimental Series 2 was used as the main experiment to investigate the optimum gaze-triggering dwell time by using a comprehensive evaluation of the task completion rate, reaction time, and NASA-TLX (Task Load Index). In Experimental Series 1, the optimal element size was determined to be 256 × 256p x 2. The conclusion of Experimental Series 2 was that when the dwell time is set to 600 ms, the efficiency of the interface is the highest, and the task load of subjects is minimal as well. Finally, the results of Experiment Series 1 and 2 have positive effects on improving the usability of the interface. The optimal control size and the optimal dwell time obtained from the experiments have certain reference and application value for interface design and software development of the ECI system.