Target Recognition for the Automatically Targeting Variable Rate Sprayer
Abstract
In order to quickly and accurately realize target recognition when the crop demanding, adopt green strength of RGB‘s method to identify the crop from the background of elaphic, the recognition accuracy was more than 98%. Use density DBSCAN fuzzy clustering algorithm, by means of setting two parameters ε and MinN, in the density DBSCAN fuzzy clustering algorithm, to determine a cluster gather on the each target, by means of judging the existence of the clustering gather or not to determine the target. With the help of image division technology, calculate the area ratio of the target and calculate the need of quantity of spraying. By means of calculating the centre point of each clustering gather(target), realize to control sprayer motion trail, realize accurately targeting and variable rate spraying pesticide (Fertilizer) when the crop sowing in line at seeding.
Origin | Files produced by the author(s) |
---|