Spatiotemporal Analysis of Event-related Current Density Reveals Dissociable Effects of Arousal and Valence on Emotional Picture Processing.

BACKGROUND: The processing of emotional visual stimulation involves the processing of emotional and visuoperceptual information. It is not completely revealed how the valence and arousal affect these two aspects. The objective was to investigate the effects of valence and arousal on spatiotemporal characteristics of cortical information processing using distributed source imaging of event-related current density (ERCD). METHODS: Electroencephalograms (64 channels) were recorded from 19 healthy men while presenting affective pictures. Distributed source localization analysis was adopted to obtain the spatiotemporal pattern of ERCD on cortical surface in response to emotional visual stimulation. A nonparametric cluster-based permutation test was used to find meaningful time and space without prior knowledge. RESULTS: Significant changes of ERCD in 400-800 ms among positive, negative, and neutral emotional conditions were found in left posterior cingulate cortex (PCC) and right inferior temporal cortex (ITC). In the PCC, the stimuli with higher arousal levels showed more negative ERCD than neutral stimuli. In the ITC, the ERCD for negative stimuli was significantly more negative than those of positive and neutral ones. CONCLUSION: Arousal and valence had strong influence on memory encoding and visual analysis at late period. The location and time showing significant change in neural activity according to arousal and valence would provide valuable information for understanding the changes of cortical function by neuropsychiatric disorders.

New Dark Area Sensitive Tone Mapping for Deep Learning Based Traffic Sign Recognition.

In this paper, we propose a new Intelligent Traffic Sign Recognition (ITSR) system with illumination preprocessing capability. Our proposed Dark Area Sensitive Tone Mapping (DASTM) technique can enhance the illumination of only dark regions of an image with little impact on bright regions. We used this technique as a pre-processing module for our new traffic sign recognition system. We combined DASTM with a TS detector, an optimized version of YOLOv3 for the detection of three classes of traffic signs. We trained ITSR on a dataset of Korean traffic signs with prohibitory, mandatory, and danger classes. We achieved Mean Average Precision (MAP) value of 90.07% (previous best result was 86.61%) on challenging Korean Traffic Sign Detection (KTSD) dataset and 100% on German Traffic Sign Detection Benchmark (GTSDB). Result comparisons of ITSR with latest D-Patches, TS detector, and YOLOv3 show that our new ITSR significantly outperforms in recognition performance.