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

A detection method of Cordyceps sinensis based on self-encoding feature learning

A Cordyceps sinensis, feature learning technology, applied in the direction of instruments, computing, character and pattern recognition, etc., can solve the problems that have been published in the literature and have no detection methods for Cordyceps sinensis

Active Publication Date: 2020-12-25
NANJING UNIV OF SCI & TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Patent searches and searches of various scientific and technological literature at home and abroad show that no Cordyceps detection method based on self-encoded feature learning has been published in the literature

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 detection method of Cordyceps sinensis based on self-encoding feature learning
  • A detection method of Cordyceps sinensis based on self-encoding feature learning
  • A detection method of Cordyceps sinensis based on self-encoding feature learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0011] The present invention is based on the Cordyceps sinensis detection method of self-encoding feature learning, and its steps are:

[0012] 1. Preparation of samples

[0013] 1) collecting a series of images containing Cordyceps sinensis;

[0014] 2) Extract the sub-blocks containing Cordyceps sinensis and the sub-blocks containing the background, the extraction ratio is 1:10, and scale them to the same size;

[0015] 3) Classify the samples according to the positive and negative sample categories and number them.

[0016] 2. Classification model training

[0017] 1) Shuffle the order of the sample library, use a two-layer stacked self-encoder network for training, and obtain the model parameters after training;

[0018] 2) Encode the sample library using the trained model,

[0019] 3) At the same time, the code and the sample category label are brought into the nonlinear kernel support vector machine for training to obtain the classification model parameters, wherein ...

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 method for detecting Cordyceps sinensis based on self-encoding feature learning. (1) Collect a series of images containing Cordyceps sinensis; (2) Extract Cordyceps sinensis and other backgrounds in the images, and make positive and negative samples of the same size; ( 3) Train the self-encoding model through the extracted sample images to obtain the parameters of the self-encoding model; (4) encode the samples through the self-encoding model; (5) use the non-linear support vector machine for classification training on the obtained encoding and sample categories, Obtain the parameters of the classification model; (6) collect the image of Cordyceps sinensis to be detected, and divide it into blocks at multiple scales; (7) encode each image with a self-encoding model, and use a nonlinear support vector machine to classify and record (8) Eliminate the detected overlapping areas and mark all non-overlapping areas on the image to be detected. The method of the invention can automatically detect the Cordyceps sinensis in the environment under complex background.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence application, in particular to a method for detecting Cordyceps sinensis based on self-encoding feature learning. Background technique [0002] Cordyceps sinensis, also known as Cordyceps sinensis, is a precious nourishing medicinal material commonly used by Chinese folks. Its nutritional content is higher than that of ginseng. It can be used as medicine or edible. It is an excellent delicacy with high nutritional value. Cordyceps sinensis is mainly produced in the upper reaches of the Jinsha River, Lancang River, and Nujiang River. It reaches Liangshan in Sichuan Province in the east, Pulan County in Tibet in the west, Minshan in Gansu Province in the north, and the Himalayas and Yulong Snow Mountain in Yunnan Province in the south. The best Cordyceps grows on sunny and humid hillsides, meadows, and bushes with soft and fertile soil at an altitude of about 3,000 to 5,500 meters, a...

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
CPCG06F18/2411G06F18/214
Inventor 张浩峰周玲莉刘世钰
Owner NANJING UNIV OF SCI & TECH