Division of Area of Fixation Interest for Real Vehicle Driving Tests

Abstract

The area of interest (AOI) reflects the degree of attention of a driver while driving. The division of AOI is visual characteristic analysis required in both real vehicle tests and simulated driving scenarios. Some key eye tracking parameters and their transformations can only be obtained after the division of AOI. In this study, 9 experienced and 7 novice drivers participated in real vehicle driving tests. They were asked to drive along a freeway section and a highway section, wearing the Dikablis eye tracking device. On average, 8132 fixation points for each driver were extracted. After coordinate conversion, the MSAP (Mean Shift Affinity Propagation) method is proposed to classify the distribution of fixation points into a circle type and a rectangle type. Experienced drivers’ fixation behavior falls into the circle type, in which fixation points are concentrated. Novice drivers’ fixation points, which are decentralized, are illustrated in the rectangle type. In the clustering algorithm, the damping coefficient λ determines the algorithm convergence, and the deviation parameter p mainly affects the number of clusters, where larger p values generate more clusters. This study not only provides the cluster type and cluster counts, but also presents the borderlines for each cluster. The findings provide significant contribution to eye tracking research.

Evaluation of apparent effectiveness of safety sign group in underground cavern construction

“…Count the occurrence number Ur of the same number r in the sequence of fixation points, and calculate the retrospective saccade number of fixation points [24] :…”

Exploring the Visual Attention Mechanism of Long-Distance Driving in an Underground Construction Cavern: Eye-Tracking and Simulated Driving

“…where TST mi is the total saccade time of the ith subject in the mth scene, TFT mi is the total fixation time of the ith subject in the mth scene, and DVP mi is the duration of visual perception of the ith subject through visual behavior in the mth scene, which is the sum of the durations of the gaze and saccade behavior. Area of interest fixation duration (AFD): AFD represents the mean driver fixation duration on a single AOI [55]. It determines the distribution of drivers' attention on a single visual cue.…”

A Survey of Eye Tracking in Automobile and Aviation Studies: Implications for Eye-Tracking Studies in Marine Operations

“…For example, in a typical K-means clustering algorithm method, the initial parameters are usually central coordinates for datasets. Hyperparameters, such as the cluster number k and the maximum iteration number n, can also be initialized; Xu et al [75] initialized the successive iterations for a given cluster center and derivation parameter P in his research using the improved affinity propagation algorithm. These numbers can largely influence the clustering result and the performance of the clustering.…”

“…Validation: The clustering result needs to be validated because it is not always satisfactory, e.g., not consistent with research expectation. If the clustering results do not align with the requirements that the defined AOI number and AOI distribution show consistency with driving common sense [75], the solution usually is to change the initial value or some other parameters, and repeat the clustering procedure until satisfactory results are achieved [75]. Fig.…”

“…Typical AOI setting definitions determined by (a) experts (adapted from[32]), (b) generation from stimulus (adapted from[68]), (c) attention map (adapted from[40]), and (d) clustering algorithm (adapted from[75]). …”