Drug identification system based on deep learning and identification method thereof

A technology of deep learning and identification method, which is applied in the field of drug identification based on deep learning, can solve the problems of similar drug packaging, confusing drugs and other problems, and achieve the effect of avoiding instability and blind spots.

Inactive Publication Date: 2018-01-05
张晨
View PDF0 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the automated method greatly reduces the error rate, there are still situations such as similar drug packaging or highly confusing drug addition. [1][2]

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
  • Drug identification system based on deep learning and identification method thereof
  • Drug identification system based on deep learning and identification method thereof
  • Drug identification system based on deep learning and identification method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] The whole method is divided into the following three steps:

[0015] 1. Preprocessing

[0016] 1.1 Obtaining the grayscale image

[0017] In the preprocessing stage, since the true color image contains a large amount of information, it will seriously affect the speed of image processing. Therefore, we first perform grayscale processing on the image. A true color image can also be called an RGB image. Each pixel on the image has three components of R, G, and B. This paper adopts the weighted average method, that is, assigns different weights to the three components of each pixel. Then calculate their weighted average with the following formula

[0018]

[0019] In the formula: WR, WG, and WB are the weights of R, G, and B, respectively. After many tests and verifications, in the existing picture library of drug packaging in the hospital pharmacy, we found that in the complex image background, when WR = 0.291 , WG=0.592, WB=0.117 is the most reasonable grayscale ima...

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 is a drug recognition system and its recognition method based on deep learning, including: 1. Preprocessing of drug image training samples, test samples, and target images. Including obtaining grayscale image, edge detection and frame processing. 2. Construct a model unit for establishing a drug recognition model based on deep learning. Including setting the initial parameters of each network layer and classifier of CNN, and inputting the above preprocessed image into CNN to obtain the features we need. 3. The identification unit is used to input the drug with identification, and then the above-mentioned established drug identification model based on deep learning obtains the convolution features of the image in the deep learning model, and cascades the obtained convolution features into the input The classifier finally outputs the recognition result. 4. Identify the hardware part of the system. It has a more accurate recognition rate and speed in various complex backgrounds and drug information, which can better reduce the error rate in drug application and the subsequent individual drug safety application platform.

Description

Technical field [0001] The invention belongs to the technical fields of drug management, computer vision and deep learning, and in particular relates to a drug identification method based on deep learning. Background technique [0002] With the rapid development of the pharmaceutical industry, there are many kinds of medicines on the market, and there are more than 1,500 kinds of medicines in most hospitals. In hospitals where medicines are still picked and dispensed manually, there are misinterpretations of prescriptions transmitted by the hospital information system (HIS) and wrong addition of medicines with similar or confusing medicines, which increases the workload of pharmacists and errors in dispensing medicines. On the other hand, the current popular automatic dispensing machine mainly works by receiving the information of settled outpatient prescriptions from the hospital HIS, automatically sending the medicines stored in it into the smart medicine basket according ...

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): G06F19/00G06K9/62G06N3/08
Inventor 张晨杨帆
Owner 张晨
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