A Multi-Feature Fruit Sorting System Using FPGA and Machine Vision
Fruit automated sorting systems can save the labor for repetitive work because they make the work more objectified and time-ef-ficient without subjective influences.However,the current high costs of these systems have prevented them from wider applications.There-fore,an edge-based fruit sorting system that is money-saving is designed on an FPGA platform using multi-features and machine vision.The multi-feature recognition model of ZWM is utilized to define multi-feature and recognition pattern vector sets for the generation of a multi-feature fruit recognition system capable of fruit sorting and grading.These vector data are then converted to rudder parameters to control the robotic arm to automatically sort fruits,recognizing them and carrying them to the designated regions.Remote control of fruit sorting is also available on mobile phones by using this cloud-connected system.This multi-feature fruit sorting system is not only cost-ef-fective but efficient in fruit sorting and fusion grading.It marks new progress in upgrading intelligent fruit sorting systems.
FPGAmachine visionmulti-feature recognition patternsorting systemimage recognitionrobotic armremote control