Design and implementation of autonomous flower harvester using image processing

Abstract

In this paper to design and implementation of fully autonomous system which can harvest flowers. The flowers will be able to cut from the plant by using the device proposed in this paper in a perfect condition which will take lesser time for harvesting as compared to the manual harvesting by humans. In recent days, automated flower harvesting is available only for large flowers like tulip. Thus there will be a requirement for harvesting smaller flowers like rose. This procedure will solve this problem and also it is cost efficient then manual harvesting. In this procedure the watershed algorithm is used to detect the flowers. By the histogram distance calculation, the detected flower is compared to the flowers which are already present in the database. If the detected flower matches 70-80%, then there will be the calculation of centroid of the flower and the distance from the centroid at which the stem is to be cut. The robotic arm is provided that will cut the matched flower when the signal has received to it via microcontroller. The project has to establish a cost effective harvesting systems for agricultural purpose.