Deep learning-based automatic detection method and system for multiple growth periods of rice ears

A technology of deep learning and automatic detection, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as limited detection and inaccuracy

Pending Publication Date: 2021-03-30
SOUTH CHINA AGRI UNIV
View PDF4 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides an automatic detection method and system based on deep learning for multiple rice ear development stages in order to overcome the technical defects of existing rice ear development stage detection methods that are limited by windy weather or inaccurate detection in complex scenes

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
  • Deep learning-based automatic detection method and system for multiple growth periods of rice ears
  • Deep learning-based automatic detection method and system for multiple growth periods of rice ears
  • Deep learning-based automatic detection method and system for multiple growth periods of rice ears

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Such as figure 1 As shown, a deep learning-based automatic detection method for multiple developmental stages of rice ears, including the following steps:

[0050] S1: Construct and train a target detection network based on deep learning to obtain a target detection model;

[0051] S2: Online continuous collection of n-day image data in the monitoring area;

[0052] S3: The image data is used as the input of the target detection model, and the number of various rice ears with different maturity levels in each image is detected;

[0053] S4: Calculate the average value of the number of various rice ears in all image data collected in the i-th day as the average number of rice ears of each type on that day;

[0054] S5: Calculate the density of various rice ears in the image according to the area of ​​the monitoring area and the daily average number of rice ears of each type;

[0055] S6: Judging whether the density of various rice ears has increased significantly and ...

Embodiment 2

[0066] More specifically, on the basis of Example 1, in order to solve the problem that other image sequence-based automatic detection methods for rice ear development are only applicable to a certain key development stage of rice ears, a deep learning-based method is provided. The automatic detection method of multiple developmental stages of rice ears can use ground monitoring equipment to acquire rice image sequences in real time to accurately detect which developmental stage rice is in, and has good applicability for different varieties of rice in different scenarios. Specifically:

[0067] First, based on the crop big data collection platform, the image data of rice from earing to harvest is obtained. Preferably, the images are taken at an angle close to the vertical, and the collected real-time front and bottom images are taken as the research object. Note that it is necessary to measure the actual area S of the monitoring area of ​​the camera on the spot. It is also pos...

Embodiment 3

[0074] More specifically, the present invention will be described in further detail below with reference to the examples and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0075] The camera deployment strategy of the present invention is as attached figure 2 , Jinnong silk seedling rice is planted in a square field of about 40m×40m, with 2 to 3 seedlings in each hole, and the hole interval is about 15-18cm. In the experiment, four identical cameras were arranged in the four corners of the field for rice monitoring. The cameras used Hikvision DS-2DC7423IW-A series smart ball cameras, fixed at a height of 2.5m from the ground, and the field of view of the lenses was the same. The actual area of ​​the monitoring area is S.

[0076] The present invention is based on deep learning to realize the automatic detection method of multiple developmental stages of rice ears, and the overall technical flow chart of the method is as attached...

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 provides a deep learning-based automatic detection method for multiple growth periods of rice ears. The method comprises the following steps: establishing a target detection model; continuously acquiring image data of a monitoring area on line, and detecting the number of various rice ears with different maturity in each image through a target detection model; calculating the averagenumber of various rice ears in the ith day; calculating the density of various rice ears of the image; judging whether the density of various rice ears is obviously increased to reach a preset threshold value or not, and if so, judging that the rice ears enter corresponding development periods; and otherwise, recalculating the average number of various rice ears on the other day. According to theprovided deep learning-based rice spike multi-development-period automatic detection method, images in different development periods of image data are automatically extracted by establishing a targetdetection model, target detection of rice spikes is realized, it is not required to segment rice spike areas, the detection rate is high, the practicality is high, the method is not influenced by strong wind weather or complex scenes, and finally, automatic detection of multiple development periods of rice ears is realized.

Description

technical field [0001] The present invention relates to the technical field of detection of multi-developmental stages of rice ears, and more specifically, to an automatic detection method and system for multi-developmental stages of rice ears based on deep learning. Background technique [0002] Accurate observation of the key developmental stages of rice panicle can effectively guide precise control, which is of great significance to the realization of high-quality and high-yield rice. Rice panicles at different growth and development stages from heading to harvest show different morphological characteristics. For example, in the heading and flowering stage of rice, the spikes are dotted with small white spikelets; in the milky stage of rice, there are no spikelets and the spikes are bent and drooping, and some are divergent; The ears appear and the chaff turns yellow. According to the "Agricultural Meteorological Observation Standards", when 10% of the rice in the obser...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06V10/44G06V10/462G06V2201/07G06F18/214
Inventor 肖德琴张远琴杨文涛黄一桂冯健昭林探宇欧周才杨兴召吴彻
Owner SOUTH CHINA AGRI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap