Low fruit picking device based on machine vision

Abstract

The invention provides a low fruit picking device based on machine vision, which comprises a negative pressure mechanism, a detection mechanism, a mobile base, a vision module, a shearing mechanism and a control module. Under analysis and manipulation of the control module, the device obtains the position information of low fruits by the vision module. The negative pressure mechanism forms a selective adsorption region to absorb fruits on an adsorption pore net according to the position information of the fruits. The detection mechanism is bound to the shearing mechanism. After the fruit is adsorbed and fixed on the adsorption pore net, the detection mechanism moves forward on the mobile base and detects the fruit to the stalk. The shearing mechanism cuts the stem of the fruit, and the adsorption force is removed, so that the fruit naturally falls into a collection basket on the base. The low fruit picking device based on machine vision has a wide application range, high automation degree, little damages to fruits, saves labor force and improves the picking efficiency.

Description

technical field [0001] The invention relates to a fruit picking device, in particular to a low fruit picking device based on machine vision. Background technique [0002] In terms of agricultural picking, picking devices are one of the main means of developing modern agricultural production, and in the picking process, precise positioning and no damage to fruits are one of the key technologies for successful completion. Therefore, the visual positioning system in the picking device must be able to identify and locate the fruit accurately and effectively. Especially, the low-short fruit usually has a short plant, and people often need to bend their knees or bend over to pick the fruit. Picking the fruit like this not only makes the picker very hard, but also the picking speed is slow and the efficiency is low. time, it will greatly increase labor costs. [0003] However, in today's technological development, there are few research results on picking devices, and due to the ...

AI Technical Summary

Problems solved by technology

The existing visual recognition method uses the singular value decomposition method and phase warping technology to extract the color and texture features of the fruit, and then uses the least squares method to fit the fruit texture features. This method has low recognition accuracy and is relatively complicated. Slow speed, not conducive to daily picking

Images