Abstract: The invention discloses an object location method, device and storage medium based on image segmentation, the object location method comprises: collecting and labeling training images to obtain a trained database; designing a fully convolutional neural network (FCNN); training the FCNN to obtain a target neural network, by inputting the trained database into the FCNN; labeling and locating object images, based on the target neural network. The method is using the training samples collected in the application scenario to train the FCNN, so it can obtain an optimized FCNN and achieve higher robustness and segmentation accuracy. Particularly, the object segmentation method in the embodiment can perform high-precision segmentation on a plurality of overlapping envelope regions when processing envelopes in a logistics system, and an envelope on the top layer is accurate singulation, allowing the robot to grab only one envelope each time, greatly improving the accuracy and experience of logistics sorting.

Abstract: The invention provides a controlling method of an unattended retail store and device thereof, and computer readable storage medium. The controlling method of the unattended retail store comprises the steps: obtaining a shelf data and a stacking data on a shelf via a visual identification device; determining whether to replenish goods based on the shelf data and the stacking data; controlling the robot to place the goods onto the corresponding shelf when it is determined that replenishment is required. The invention improves the efficiency and automation of the controlling of the unattended retail store.

Abstract: The invention discloses an object location method, device and storage medium based on image segmentation, the object location method comprises: collecting and labeling training images to obtain a trained database; designing a fully convolutional neural network (FCNN); training the FCNN to obtain a target neural network, by inputting the trained database into the FCNN; labeling and locating object images, based on the target neural network. The method is using the training samples collected in the application scenario to train the FCNN, so it can obtain an optimized FCNN and achieve higher robustness and segmentation accuracy. Particularly, the object segmentation method in the embodiment can perform high-precision segmentation on a plurality of overlapping envelope regions when processing envelopes in a logistics system, and an envelope on the top layer is accurate singulation, allowing the robot to grab only one envelope each time, greatly improving the accuracy and experience of logistics sorting.