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

A seed identification method based on multi-scale feature fusion and extreme learning machine

An extreme learning machine and multi-scale feature technology, which is applied in character and pattern recognition, computer components, instruments, etc., can solve the problem of small number of recognition models

Active Publication Date: 2018-12-25
FUZHOU UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these studies are all based on a single-feature recognition model, the number of recognition is small, and special hardware equipment is required.

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
  • A seed identification method based on multi-scale feature fusion and extreme learning machine
  • A seed identification method based on multi-scale feature fusion and extreme learning machine
  • A seed identification method based on multi-scale feature fusion and extreme learning machine

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057] The technical solution of the present invention will be specifically described below in conjunction with the accompanying drawings.

[0058] The present invention provides a seed identification method based on multi-scale feature fusion and extreme learning machine, such as figure 1 As shown, follow the steps below to achieve:

[0059] Step S1: Multi-scale fusion feature extraction is performed on the multi-scale local HOG features and globally distributed HSV features of the sample seed image in the image dataset, and the multi-scale fusion feature information of the sample seed image is obtained;

[0060] Step S2: use the cross-validation training of the ELM (Extreme Learning Machine) extreme learning machine to perform feature training on the multi-scale fusion feature information of the sample seed image, and establish an integrated classification model;

[0061] Step S3: Classify, detect and identify the seed image to be tested by adopting the sliding window mecha...

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 present invention relates to a seed identification method based on multi-scale feature fusion and extreme learning machine. By merging and extracting multi-scale local HOG features and globally distributed HSV features of sample seed images, the ELM extreme learning machine is used for feature training to obtain An integrated classification model is developed, and the sliding window mechanism is used to extract the multi-scale fusion feature of the seed image to be tested, and the result is input into the model, and the weighted voting of the classification result is carried out to obtain the seed image to be tested. Image classification information. A seed identification method based on multi-scale feature fusion and extreme learning machine proposed by the present invention is simple and flexible, requires simple equipment, and has strong practicability.

Description

technical field [0001] The invention relates to the field of seed recognition based on machine vision, in particular to a seed recognition method based on multi-scale feature fusion and extreme learning machine. Background technique [0002] Seeds are the most basic means of production in agricultural production and the basis of all agricultural production. Accurate and rapid identification of seeds has important guiding significance for production. Commonly used methods for seed identification at home and abroad mainly include fluorescence scanning identification method, chemical identification method and electrophoresis identification method. These methods require strong professional background knowledge and long identification time and will cause certain damage to the seeds. Therefore, how to non-destructively and quickly identify crop seed categories is an important research direction. In recent years, with the rapid development of machine learning, many scholars at h...

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): G06K9/62
CPCG06F18/2413G06F18/214
Inventor 柯逍杜明智周铭柯
Owner FUZHOU UNIV