Rotary adjustable binocular vision target recognition and positioning device and application thereof in agricultural fruit harvesting machinery

Abstract

The invention discloses a rotary adjustable binocular vision target recognition and positioning device and application thereof in agricultural fruit harvesting machinery. The target recognition and positioning device comprises a support, cameras and an image processing system, wherein the support comprises two longitudinal guide rails and two transverse guide rails; the transverse guide rails and the longitudinal guide rails are in sliding connection to realize movement of the cameras in Z-axis direction; the cameras are two cameras which are fixedly arranged on the support through a pedestal and are symmetrically arranged left and right; the pedestal can rotate around Y-axis; the pedestal and the transverse guide rails are in sliding connection to realize movement of the cameras in X-axis direction. The target recognition and positioning device can realize multi-degree-of-freedom movement of binocular vision three-dimensional cameras in a three-dimensional space, realize binocular vision multi-angle recognition and positioning and optimize the recognition and positioning effects; the device is applied to the agricultural fruit harvesting machinery, can adapt to various agricultural fruits and has certain universality.

Description

technical field [0001] The invention relates to the field of agricultural picking, in particular to a rotating adjustable binocular vision target recognition and positioning device and its application in agricultural fruit picking machinery. Background technique [0002] Most of the existing agricultural fruit picking machines are manual picking or manual picking with simple mechanisms, which are usually only suitable for small-scale operations. There are problems such as high labor costs, low picking efficiency, and low degree of automation. It is difficult to adapt to large-scale planting of agricultural fruits and picking requirements. To improve the intelligence level of agricultural fruit picking machinery, it is first necessary to improve the efficiency of fruit identification and positioning. The shape, size, color and growth position of fruits growing on plants or trees are very different, and the fruits are always randomly distributed and grown. However, the existi...

AI Technical Summary

Problems solved by technology

The shape, size, color and growth position of fruits growing on plants or trees are very different, and the fruits are always randomly distributed and grown. However, the existing agricultural fruit picking machines generally use fixed target recognition and positioning devices, which can only The image information of the fruit is obtained within a certain angle range, and the recognition range of the naturally growing fruit is limited

Existing target recognition and positioning devices mainly include: monocular and binocular recognition systems fixed on the terminal actuator of the manipulator or on the manipulator. identify

Moreover, it is difficult for the fixed-position camera in the existing target recognition and positioning device to identify many small fruits on the length of a plant at one time, and when the vertical drooping growth of the plant requires proper adjustment of the viewing angle, it cannot effectively identify

Therefore, the existing device cannot allow the picking mechanism to pick all the fruits once without damaging the fruits.

For the picking of this special growing small forest fruit, the vision system of the picking machine needs to be able to continuously adjust the visual range and angle to obtain the best fruit target recognition and positioning environment and the angle of view that is least affected by light. However, the existing agricultural fruit picking Machinery cannot meet this requirement

Images