Method for distinguishing medicinal plants based on image identification of random distribution of identification fibers

A random distribution, image recognition technology, applied in the direction of character and pattern recognition, computer parts, instruments, etc., can solve the problems of indistinguishable identification, etc., and achieve the effect of solving dyeing counterfeiting and improving sensitivity

Inactive Publication Date: 2018-09-21
徐飞
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Purpose of the invention: This invention applies image recognition technology to the identification of Chinese herbal medicines for the first time, and creatively uses image recognition to identify the random distribution of fibers in medicinal plants to distinguish their characteristics, which can effectively solve the existing visual problems in the market such as dyeing and adulteration of Chinese herbal medicines. Biomimetic systems and other technologies cannot distinguish and identify problems

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
  • Method for distinguishing medicinal plants based on image identification of random distribution of identification fibers
  • Method for distinguishing medicinal plants based on image identification of random distribution of identification fibers
  • Method for distinguishing medicinal plants based on image identification of random distribution of identification fibers

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] Embodiment 1 Image background elimination and segmentation of fiber image of medicinal herb decoction pieces

[0028] The present invention adopts an adaptive threshold segmentation method. The image background removal is to use the gray histogram of the R component of the image of the decoction pieces to remove the black background in the image of the decoction pieces. After the image background is removed, it is necessary to segment the chrysanthemum heart and outer skin in the decoction pieces. In order to effectively improve the precision of image segmentation, the present invention adopts a fuzzy clustering method based on fuzzy set theory. First use the contour tracking technology to trace the outermost contour of the decoction pieces, and perform direction projection within the contour, and delineate the approximate range of the chrysanthemum heart according to the projection information. figure 2 yes figure 1 The R component gray histogram of the decoction p...

Embodiment 2

[0029] Example 2 Extraction of Radix Astragalus Decoction Pieces Labeled Fiber - Image of Chrysanthemum Heart

[0030] The extraction of the identified fiber of Astragalus decoction pieces—the image of chrysanthemum heart mainly includes the following three steps:

[0031] 1. Extract the solid area of ​​the chrysanthemum heart (including the obvious cambium pattern and radial texture) from the whole image;

[0032] 2. Separate the effective judging parts of the chrysanthemum heart from the outer skin and other structures (that is, the cambium pattern and radial texture);

[0033] 3. Extract the identification fiber chrysanthemum heart.

[0034] Overview of the extraction method: first use the contour tracking technology to trace the outermost contour of the decoction pieces, and perform direction projection within the contour, and delineate the approximate range of the chrysanthemum heart according to the projection information. In order to determine the richness and texture...

Embodiment 3

[0035] The feature extraction of embodiment 3 chrysanthemum heart

[0036] Through the on-the-spot investigation of the Chinese herbal medicine market, it was found that the size of the decoction pieces of Astragalus membranaceus has a great influence on the condition of the chrysanthemum heart. And there are also big differences in the chrysanthemum hearts of the same level and different sample sizes. Therefore, the present invention selects the image feature pixel gray moment feature of the sample size and shape of the decoction pieces as the chrysanthemum heart image feature, and simultaneously improves the robustness of the computer vision detection system. The (m+n) order geometric moment of the two-dimensional function of the digital image is defined as:

[0037]

[0038] The (m+n) order central moment of a two-dimensional function is defined as:

[0039]

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 method for distinguishing features of medicinal plants based on image identification of random distribution of identification fibers. By utilizing an image identification technology, the features of shapes, textures and surface vectors of traditional Chinese medicinal materials are effectively obtained; and then according to the random distribution of the identification fibers in the medicinal plants, different textures are formed, and serve as an important parameter for distinguishing the authenticity and quality of the medicinal plants. The technology is relatively high in objectivity and accuracy, can be accurately quantized, and is slightly influenced by sensory difference and environment monitoring. The method is simple, easy and quick, is relatively high in operability, and makes up for the limitation of a visible range of human eyes to a great extent.

Description

technical field [0001] The invention relates to the field of identification of medicinal materials, in particular to a method for distinguishing the characteristics of medicinal plants by using image recognition to identify the random distribution of fibers. Background technique [0002] The traditional character identification of Chinese herbal medicines has the following disadvantages: [0003] 1. Traditional trait identification relies on sensory evaluation by experts in traditional Chinese medicine identification, which is especially subjective and cannot be accurately quantified and objectively measured. It can only be roughly estimated and briefly described in language, and has poor operability in mass industrial production. Inevitably affected by sensory differences and monitoring environment. [0004] 2. The visual bionic system currently on the market mainly extracts the color features of Chinese medicinal materials, but does not extract and analyze their shape 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
IPC IPC(8): G06K9/62G06K9/46
CPCG06V10/44G06V10/758G06F18/23G06F18/214G06F18/2411
Inventor 徐飞杜文嘉陆彩尤敏
Owner 徐飞
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