The “Mask Effect” on the tips that customers leave restaurant servers

Masks have become the custom among restaurant workers and bartenders as a form of protection against COVID-19. Yet, given the rapid introduction of masks to the uniforms of restaurant servers there is a dearth of extant scholarship that has explored the effects of face coverings on customers’ behaviors. In response, this research offers a preliminary test of the effect of server masks on a common consumer behavior in the full-service restaurant industry—tipping. We review theoretical and empirical reasons suggesting that a mask may have a negative effect on customers’ tipping practices. Potential mask effects on tipping are then explored with a survey experiment that was administered to a large and diverse sample of Amazon Mechanical Turk “workers.” Our results suggest that wearing a mask is not likely to, on average, have a meaningful effect on how much restaurant customers tip their servers. However, we do observe a negative indirect effect of a mask on customer reported tip amount through diminished perceptions of a hypothetical servers’ friendliness. This effect was found to be attenuated among those who are altruistically motivated to tip servers as a way to help them financially. The implications of our results and directions for future research are discussed.     

采用深度卷积神经网络方法的毫米波图像目标检测

通过收集大量的毫米波图像并建立相应的人体数据集进行检测，提出基于Faster R-CNN深度学习的方法检测隐藏于人体上的危险物品。该方法将区域建议网络和VGG19训练卷积神经网络模型相结合，构建了面向毫米波图像目标检测的深度卷积神经网络。为了提高毫米波图像的处理能力，采用Caffe深度学习框架在图形处理单元上进行训练和测试。实验结果证明了基于Faster R-CNN深度卷积神经网络的目标检测方法能有效检测毫米波图像中的危险物品，并且目标检测的平均准确率约94%，检测速度约为6 frame/s，对毫米波安检系统的智能化发展有着极其重要的参考价值。