Development of Application Sales Support for Supermarket Using Artificial Intelligence and Customer Relationship Management

This research presents a study of the shopping behaviors. Shops generally installed intelligent surveillance camera we propose to use benefit from existing solutions. In training, we used only interested frames in order to generated frequency We explicitly reformulate the layers as learning residual functions with reference to the layer inputs represent frequencies with Heatmap diagram. During testing, since all the image’s sequences are input into the system, we discovered frequencies of images cannot identify as person as we expected. We proposed to improve this logic during implementation by remove empty space area out of generated frames with by clustering frequencies layers of data before representing with Heatmap diagram.