Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Plant nematode detection method and system based on multi-feature combination based on deep learning

A plant nematode and deep learning technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as inconspicuousness, manpower limitations, and identification difficulties, and achieve high detection accuracy, high accuracy, and fast speed Effect

Active Publication Date: 2022-08-02
NINGBO INST OF MATERIALS TECH & ENG CHINESE ACADEMY OF SCI
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the small size of the nematode, the key feature points of the nematode such as the head and tail are smaller and less obvious than the nematode body. Sometimes it is necessary to jointly detect multiple features to distinguish the type of B. xylophilus, which is very difficult to identify.
[0003] The traditional detection method is based on the shape of nematodes, mainly relying on accumulated experience to identify familiar groups, which is greatly limited by manpower

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
  • Plant nematode detection method and system based on multi-feature combination based on deep learning
  • Plant nematode detection method and system based on multi-feature combination based on deep learning
  • Plant nematode detection method and system based on multi-feature combination based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] like figure 1 As shown, a multi-feature combined plant nematode detection method based on deep learning, through single-point feature detection, double-point feature combined detection and three-point feature combined detection, to achieve the detection of plant nematodes (B. xylophilus and B. xylophilus). ) accurate detection. Specifically include the following steps:

[0057] Step S0, establish a plant nematode image data set;

[0058] A series of images of B. xylophilus and B. xylophilus were collected to build a plant nematode image dataset.

[0059] Step S1, using Fast-RCNN neural network to extract key features of plant nematodes, including nematode head features, female nematode tail features and male nematode tail features. like figure 2 shown, including the following steps:

[0060] S11, establishing a plant nematode image training data set;

[0061] The plant nematode image data set is divided into a pine wood nematode image group and a pine wood nemato...

Embodiment 2

[0081] A multi-feature combined plant nematode detection system based on deep learning, including:

[0082] A data set generation module to establish a plant nematode image data set including B. xylophilus images and B. xylophilus images;

[0083] The key feature extraction module uses the Fast-RCNN neural network to extract the key features of plant nematodes, including nematode head features, female nematode tail features and male nematode tail features;

[0084] The detection enhancement module is used to detect plant nematode species according to the key characteristics of plant nematodes, including a single-feature detection enhancement module for female nematode tail features, a nematode head feature and a female nematode tail feature dual-feature combined detection enhancement module, and a nematode head feature. An enhanced module for combined detection of three characteristics of characteristics and tail characteristics of female nematodes and tail characteristics 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 plant nematode detection method and system based on deep learning and multi-feature combination. The present invention uses the first deep learning neural network to extract the key features of plant nematodes, and respectively detects the key characteristic regions such as the head, female tail, and male tail of the plant nematodes; and uses the second deep learning neural network as the identification and detection network to achieve Single-feature detection, dual-feature joint detection and three-feature joint detection have high detection accuracy. Finally, a humanized nematode detection interface is designed, and an intelligent nematode detection system is completed. The nematode detection and identification scheme can be selected according to the degree of difficulty in distinguishing nematodes. The invention adds a mechanism for effectively extracting key features, and a method for multi-feature joint classification. The invention can accurately detect plant nematode data with small key feature points and are difficult to distinguish, and utilizes multi-feature combination to realize hierarchical detection of pine wood nematodes and pine wood nematodes.

Description

technical field [0001] The invention relates to the technical field of deep learning image intelligent detection and retrieval, in particular to a method and system for intelligent detection of plant nematodes based on a combination of deep learning and multi-features. Background technique [0002] Affected by alien species, the spread of pine wood nematode is becoming more and more serious in my country, resulting in huge economic losses every year. It is imperative to increase quarantine efforts, improve the ability to intercept the epidemic from the source, and curb the spread of pine wood nematode. However, the accurate detection of B. xylophilus among various nematodes with very similar characteristics presents a great challenge to the technicians of the detection agency. How to efficiently and accurately detect B. xylophilus has always been a very critical problem. Due to the small size of nematodes, the key features of nematodes such as head and tail are smaller and ...

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 Patents(China)
IPC IPC(8): G06V10/774G06V10/764G06K9/62G06V10/82G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/241G06F18/214
Inventor 庄佳衍顾建锋朱莹肖江剑朱屹刘阳明
Owner NINGBO INST OF MATERIALS TECH & ENG CHINESE ACADEMY OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products