Improving AR-SSVEP Recognition Accuracy Under High Ambient Brightness Through Iterative Learning

Abstract

Augmented reality-based brain-computer interface (AR-BCI) system is one of the important ways to promote BCI technology outside of the laboratory due to its portability and mobility, but its performance in real-world scenarios has not been fully studied. In the current study, we first investigated the effect of ambient brightness on AR-BCI performance. 5 different light intensities were set as experimental conditions to simulate typical brightness in real scenes, while the same steady-state visual evoked potentials (SSVEP) stimulus was displayed in the AR glass. The data analysis results showed that SSVEP can be evoked under all 5 light intensities, but the response intensity became weaker when the brightness increased. The recognition accuracies of AR-SSVEP were negatively correlated to light intensity, the highest accuracies were 89.35% with FBCCA and 83.33% with CCA under 0 lux light intensity, while they decreased to 62.53% and 49.24% under 1200 lux. To solve the accuracy loss problem in high ambient brightness, we further designed a SSVEP recognition algorithm with iterative learning capability, named ensemble online adaptive CCA (eOACCA). The main strategy is to provide initial filters for high-intensity data by iteratively learning lowlight-intensity AR-SSVEP data. The experimental results showed that the eOACCA algorithm had significant advantages under higher light intensities (>600 lux). Compared with FBCCA, the accuracy of eOACCA under 1200 lux was increased by 13.91%. In conclusion, the current study contributed to the in-depth understanding of the performance variations of AR-BCI under different lighting conditions, and was helpful in promoting the AR-BCI application in complex lighting environments.

Dataset Evaluation Method and Application for Performance Testing of SSVEP-BCI Decoding Algorithm

“…Table 7 shows that the majority of the 20 SSVEP decoding algorithm papers analyzed used more than one dataset, with only three papers [ 18 , 20 , 25 ] using self-collected data verification methods. Among the 20 papers, 40% (8/20) used only one dataset, while 80% (16/20) used no more than two datasets, and the remaining four papers used only three datasets for verification.…”

“…Therefore, score 1 is not adjusted, while score 2 is slightly reduced. Although T best is a crucial indicator that reflects the performance of the BCI system, the evaluation of the BCI system currently focuses on the accuracy and the information transfer rate [ 7 , 18 , 35 ]. Therefore, score 4 was appropriately reduced, while score 3 and score 5 were appropriately increased.…”

“…However, the majority of current algorithmic research utilizes one or two datasets to verify their performance [ 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ], which did not make full use of public data resources, and the results were limited by the distribution of data samples in individual datasets, so it was not conducive to judge the application effect of the algorithm in the actual scene through the result. This issue has two underlying causes.…”

A comparative study of stereo-dependent SSVEP targets and their impact on VR-BCI performance

“…Within each session, the four stimuli flickered at fixed frequencies as 8.5, 9.5, 11, and 12.5 Hz, positioned as shown in Figure 1A , starting with the upper object and in clockwise order. We chose these frequencies based on Wang et al ( 2016 ) and similar AR-SSVEP studies (Zhang et al, 2023 ), also with the consideration about fresh rate of VR device. We developed two methods for presenting flickering stimuli in the Unity3D program, as shown in Figures 3A , B .…”

“…However, the mature and mainstream approach in the current time mainly involves the use of CRT/LCD displays to present various planar stimuli (Wang et al, 2016 ; Liu et al, 2020 ). Since there has been research on SSVEP using an LCD display to present 3D stimuli in the past (Mun et al, 2013 ), there are emerging studies in this field that have begun exploring similar integrative applications using Virtual Reality (VR)/Augmented Reality (AR) devices (Chen et al, 2021 ; Mahmood et al, 2022 ; Zhang et al, 2023 ). Unlike traditional LCD displays, VR/AR is a binocular vision-based display device that uses parallax and immersive computer graphics effects to build an interactive environment.…”