Nondestructive detection method for withering degree indexes of black tea

A non-destructive testing and withering technology, applied in prediction, still image data retrieval, image data processing, etc., can solve the problems of low generalization performance and stability of the model, and achieve optimal generalization performance and stability, strong prediction ability, The effect of strong generalization ability

Pending Publication Date: 2020-08-11
TEA RES INST CHINESE ACAD OF AGRI SCI
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods all use color features and texture features as the input of the model, ignoring other information associated with moisture in the original image, resulting in low generalization performance and stability of the built model

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
  • Nondestructive detection method for withering degree indexes of black tea
  • Nondestructive detection method for withering degree indexes of black tea
  • Nondestructive detection method for withering degree indexes of black tea

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] The establishment process of the CNN (convolutional neural network) model that the present invention uses is as follows:

[0045] The input of the CNN model is the original picture collected by the image acquisition system, which provides more original information for the filter stage. The initial size of the input picture is 1080*1080*3. The proposed CNN model has 10 layers, including 5 Convolutional layer, 2 maximum pooling layers, 1 softmax layer, 1 fully connected layer, 1 loss function layer, the structural order is convolutional layer 1, convolutional layer 2, maximum pooling layer 1, convolutional layer Layer 3, convolutional layer 4, maximum pooling layer 2, convolutional layer 5, fully connected layer, softmax layer. Among them, the pixel size of the convolution filter in the convolutional layer is (3×3)~(13×13), the number of convolution filters is 128~512, and the convolution step is 1~3. The pixel sizes of the multilayer convolution filters are 11×11, 7×7, ...

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 relates to the technical field of tea withering degree detection methods, in particular to a nondestructive detection method for withering degree indexes of black tea. The nondestructivedetection method comprises the following steps of respectively placing tea leaves withered to different degrees under an industrial camera to shoot sample pictures, detecting withering degree indexes, preprocessing the shot sample pictures, establishing a convolutional neural network database by using withered leaf picture information and withering degree index data, classifying the withering degree indexes in the database, and establishing a black tea withering degree index quantitative prediction model according to confidence coefficients obtained by classifying the withering degree indexes. According to the nondestructive detection method, a black tea withering moisture quantitative prediction model based on the convolutional neural network confidence coefficient is established, and the model has excellent generalization performance and stability and can be used for rapid and nondestructive detection of black tea withering moisture.

Description

technical field [0001] The invention relates to the technical field of detection methods of tea withering degree, in particular to a non-destructive detection method of black tea withering degree index. Background technique [0002] Withering is the basic process of black tea processing, and its quality directly affects the quality of finished tea. As the degree of withering deepens, the fresh tea leaves will gradually wither, the leaf color will turn from bright to dark, and the green grass will gradually disappear. Usually the water content of the withered leaves is used as an indicator of whether the withering is excessive, and it is considered that the water content of the fresh leaves reaches 58%-62% when it is moderately withered. In the processing and production of black tea, the moisture content of fresh leaves is often judged by sensory experience, and sensory experience will produce errors with different judges. At the same time, it is difficult to divide them in...

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): G06T7/00G06K9/62G06N3/04G06Q10/04G06F16/55
CPCG06T7/001G06F16/55G06Q10/04G06T2207/30108G06T2207/20084G06N3/045G06F18/241G06F18/214
Inventor 董春旺安霆杨崇山刘中原杨艳芹王近近李佳江用文袁海波邓余良滑金杰
Owner TEA RES INST CHINESE ACAD OF AGRI SCI
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