Tea garden identification weeding method using convolutional neural network

Abstract

The invention relates to a tea garden identification weeding method using a convolutional neural network, and belongs to the field of mode identification. Firstly, a robot is used for collecting images with fixed angles in real time in a tea garden, the images are segmented through transverse and vertical grid lines, convolutional neural network recognition is conducted through segmented small images to obtain a classification result. Due to the fact that the tea garden environment is complex, secondary recognition of the convolutional neural network needs to be conducted on the pictures of the'mixed 'result in the obtained classification result, and the result is obtained. According to the identified result, the positions of the tea, the grass and the others are obtained, and weeding operation is automatically conducted according to the obtained positions. The labor consumption is reduced, and the weeding efficiency and the weeding quality of the tea garden are improved.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and in particular relates to a tea garden recognition and weeding method using a convolutional neural network. Background technique [0002] Tea gardens have long been harmed by weeds, which affect the absorption of fertilizers and water by tea leaves, thereby affecting the growth and quality of tea leaves. The weeding work of tea farmers has become an important part of the work of tea gardens. The traditional weeder has two wheels and a high-speed rotating motor, which pulls the weeder itself manually and quickly rotates the "weeding head" in front of the weeder to remove weeds. Although this method is slightly improved compared to the most traditional manual weeding, it still requires a person to carry a large weeding machine and go to the field to operate it in person. Working with a large machine also increases the hard work of the tea farmers. Moreover, due to the high horsepower of this...

AI Technical Summary

Problems solved by technology

Although this method is slightly improved compared to the most traditional manual weeding, it still requires a person to carry a large weeding machine and go to the field to operate it in person. Working with a large machine also increases the hard work of the tea farmers.

Moreover, due to the high horsepower of this kind of weeding machine, it is very easy to set off huge dust during the weeding process, which affects the health of the tea farmers' respiratory tract. In addition, the machine is huge, so the "weeding cutter head" is also a bit bloated. When weeding, Only some weeds that are far away from the tea tree itself can be removed, and the weeds next to the tea tree are easy to hurt the tea tree during the removal process

Images