Unlock instant, AI-driven research and patent intelligence for your innovation.

Tree fruit recognition method based on improved YOLOv3

A fruit recognition and fruit technology, applied in the field of target detection, can solve the problems of misrecognition, too much emphasis on small target recognition accuracy, no consideration of convolutional neural network depth and detection speed, etc., with fast speed, high recognition accuracy, and satisfactory The effect of real-time recognition

Active Publication Date: 2021-02-05
GUANGXI NORMAL UNIV
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the existing fruit identification method based on convolutional neural network, one shortcoming is that when identifying the fruit on the tree, it pays too much attention to the recognition accuracy of small targets, without considering the depth and detection speed of convolutional neural network; Accurate identification cannot be achieved when clustering and overlapping fruit targets on trees, and there are misidentifications

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Tree fruit recognition method based on improved YOLOv3
  • Tree fruit recognition method based on improved YOLOv3
  • Tree fruit recognition method based on improved YOLOv3

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0049] Take grapes, for example.

[0050] refer to figure 1 , a tree fruit recognition method based on improved YOLOv3, including the following steps:

[0051] S1. Image acquisition: the user uses a digital camera or other image acquisition equipment to collect images of grape vines bearing fruit. The image acquisition time includes different stages of fruit growth, early growth, middle growth and maturity. The time points for shooting images are distributed in the morning , noon, and afternoon at different times, so that the captured images include images of different time periods. Finally, the collected images are named according to the format of the Pascal VOC dataset, and three named Labeleds, PictureSets, and ResultSets are created at the same time. folder;

[0052] S2. Image preprocessing:

[0053] S2-1. Image marking: in the image collected in step S1, use the image labeling tool LabelImg to mark the grape fruit in the image, and mark the position and variety name of...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a tree fruit recognition method based on improved YOLOv3. The method comprises the following steps of S1, acquiring an image, S2, preprocessing the image, S3, setting network model parameters, S4, improving the original YOLOv3 network structure to obtain an improved YOLOv3 network structure, S5, training a network model, and S6, using the trained network model weight to carry out recognition. According to the method, large-area clustered and overlapped fruits on trees can be accurately recognized, the recognition precision is high, the speed is high, and the requirementof real-time recognition can be met.

Description

technical field [0001] The invention relates to the technical field of target detection, in particular to a tree fruit recognition method based on improved YOLOv3. Background technique [0002] my country is a big fruit producing country in the world, the fruit types are in the forefront of the world rankings, and the cultivated area ranks first in the world. Spraying, fruit picking, post-harvest sorting and other operations are important links in fruit tree production. In the current actual production in my country, limited by factors such as technology and economic investment, most of these operations are completed manually, which makes the cost of fruit production in my country high. low efficiency. In recent years, with the rapid development of information technology, many researchers have used computer vision recognition technology to assist orchard yield assessment, automatic picking, disease prevention and control, and fruit sorting applications, among which fruit ide...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/10G06V20/68G06N3/045G06F18/23213G06F18/214
Inventor 陆声链刘晓宇李帼陈文康
Owner GUANGXI NORMAL UNIV