| In recent years,recyclable resources in garbage haven't been effectively utilized,especially the packaging of drinks.It is necessary to innovate the way of recycling packaging.Today,with the rapid development of image recognition,we can use image recognition in the recovery of beverage packaging,it can recognize the beverage packaging when barcode recognition can't work.We can achieve more flexible recognition and recovery.In response to above,in this paper,an open source Zedboard development board based on ZYNQ-7000 platform is used to design a recognition and recycling system for beverage packaging,the system combines image recognition with barcode recognition.When the barcode is complete and identifiable,it can be recognized and recovered by scanning the barcode to determine the type of beverage packaging;when the barcode is incomplete and unidentifiable,it can be classified and recovered by image recognition.Based on the structure of ZYNQ-7000 platform which includes ARM and FPGA,the hardware and software co-design is carried out,image processing IP core is compiled by Vivado HLS to accelerate image processing on hardware.A high-speed image processing system is designed in Vivado.In the software part,the IP core is controlled by program;the barcode recognition function is designed,which can be recognized by scanning barcode;the beverage packaging database is established,and the image contrast algorithm is compiled.By comparing the captured image with the image in the database,the type of beverage packaging can be determined and classified.After testing the system,the system can accurately achieve the barcode recognition and image recognition.It makes up for the disadvantage that the beverage packaging can't recover when barcode can't be recognized,and achieves the expected goal.In this design,image recognition is applied to the recovery of beverage packaging.Through hardware and software co-design,the processing speed is greatly improved. |
PDF Full Text Request