Research on image classification of Chinese medicinal materials powder based on texture extraction and improved lightweight network

Yiding Wang; Senhao Tao; Yaoli Li; Shaoqing Cai; Yuan Yuan

doi:10.1117/12.2660975

2 February 2023 Research on image classification of Chinese medicinal materials powder based on texture extraction and improved lightweight network

Yiding Wang, Senhao Tao, Yaoli Li, Shaoqing Cai, Yuan Yuan

Author Affiliations +

Proceedings Volume 12462, Third International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022); 1246229 (2023) https://doi.org/10.1117/12.2660975
Event: International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022), 2022, Xi'an, China

Abstract

In recent years, a large number of mixed products, counterfeit products and adulterated products of traditional Chinese medicine powder have often appeared on the market. And because the traditional image processing algorithm cannot meet the application requirements for such complex background, large amount of data, and multi-category problem processing accuracy and speed. In view of the above problems, this paper proposes an improved method of deep convolutional neural network based on texture extraction and improved attention mechanism model. First, the MobileNetV3 network is unable to fully extract the texture information of fiber powder, and an inception-like structure is added to significantly improve the final recognition accuracy; secondly, the use of hole convolution minimizes the impact of the inception-like structure on the overall parameter quantity. Influence: finally, an improved attention mechanism module is added to the network, which significantly suppresses the noise texture in the background image. The fiber characteristics of 32 different kinds of Chinese herbal medicine powder images are used for experimental comparison. The added inception-like structure and the improved attention mechanism module increase the accuracy rate by 4.1% to 92.82%. The experiments show that the improvement method proposed in this paper is better than other classification algorithms.

Citation Download Citation

Yiding Wang, Senhao Tao, Yaoli Li, Shaoqing Cai, and Yuan Yuan "Research on image classification of Chinese medicinal materials powder based on texture extraction and improved lightweight network", Proc. SPIE 12462, Third International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022), 1246229 (2 February 2023); https://doi.org/10.1117/12.2660975

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Convolution

Medicine

Image processing

Evolutionary algorithms

Image classification

Network architectures

Image enhancement

Show All Keywords

Keywords/Phrases

Search In:

Publication Years