Design and Implementation of Parcel Sorter Using Deep Learning

Automation in industrial environment reduces the cost of the operation while increasing the overall performance system. Having an automation system in the E-commerce warehouse to sort the parcels based on their destinations or shipping method will reduce the parcel processing time significantly. To automate parcel processing in Digikala’s warehouse parcel sorter system is designed and implemented. In this system shipment method of the parcel is indicated with set of markers. computer vision system is developed to identify these markers using deep learning algorithms. The parcels are identified while they are moving on the conveyor belt in relatively high speed (1 m/s). The computer vision system is capable of processing 1.3 MP pictures in real-time with rate of 100 FPS. To sort the parcels omni wheel roller mechanism designed and utilized. To achieve best results in practical environment, gapper mechanism and pack positioning conveyor is implemented before packages reach the sorter. This system is successfully installed in the Digikala’s warehouse.