FLOWER HARVESTING ROBOT USING COMPUTER VISION

摘要

over the past decades, horticulture industry can be considered as one of the main contributors of the export sector in developing countries like Ethiopia. The horticulture industry deals with plants, mainly for food, beauty and decoration (flowers). This paper aims at providing a system that can improve outdated ways of harvesting methods. Those are mainly manual and laborious tasks which also took a lot of time. In this paper, a harvesting robot is designed and implemented using raspberry pi microcomputer to build a flower harvesting system that can identify, sort and arrange the flowers being harvested without human interaction. Here, a computer vision technology is applied to identify the types and states of the flowers. A convolution neural network image classifier is used along with deep learning networks. The harvesting activity of the machine is assisted by a 3 DOF robotic arm and additional cutting mechanism attached with cutter blade.