基于视觉的智能分拣系统的设计

北京理工大学珠海学院2020届本科生毕业设计AbstractIn recent years,with the development of e-commerce platforms,online shoppinghas become one of the daily shopping choices of many people.The development ofe-commerce has driven the development of the express delivery industry,but thetraditional express delivery industry is still using manual to sort express delivery.The advantages of sorting are high flexibility and high fault tolerance.Thedisadvantages are that it requires a lot of labor to meet the demand,cannot workcontinuously for 24 hours,and may make mistakes when facing a large number ofworkpieces.The advantage of the previous machine sorting is that it can workcontinuously for 24 hours,but it also has the increased possibility of making mistakeswhen facing a large number of workpieces,the machine itself is cumbersome,flexibleproduction cannot be achieved,and the shape of the workpiece is required.Thedesign research of this vision-based intelligent sorting system revolves around keytechnologies such as design methods,photoelectric sensors,kinematic analysis,raspberry pi,open cv-based visual recognition,and PLC control.Combining theadvantages of manual and machine sorting,the shortcomings of both are eliminated.It realizes sorting according to the characteristics of the workpiece,and realizes theability to easily modify the standard for identifying the workpiece,and realizes thefunction of flexible production.This design realized the design of a relatively low-cost intelligent sorting systemby looking for a lower-cost Mitsubishi PLC,an open-source image recognitionalgorithm,and a robot arm attitude control.This design can be used for flexibleproduction,it is easy to modify the identified workpiece,and realize intelligent andinformation production.Because the Raspberry Pi is used as the central processor,the program can be modified through remote connection to realize the remote controlof the production process.Keywords:visual recognition,raspberry pi,machine sorting PLC control目录1前言1.1选题研究意义与目的1