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

Image recognition method and device and storage medium

An image recognition and image technology, applied in the field of information processing, can solve the problems of reduced target classification and positioning accuracy, target detection methods cannot meet high precision requirements, etc., to achieve accurate target positioning, reduce target missed detection rate, and reduce leakage. The effect of detection rate

Pending Publication Date: 2021-02-02
BEIJING LINJIN SPACE AIRCRAFT SYST ENG INST
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] With the development of computer vision technology, target detection technology is more and more widely used in production and life. With the expansion of application fields, the requirements for target detection accuracy in engineering applications are getting higher and higher, especially in the aerospace field. Object detection methods cannot meet their high precision requirements
In the prior art, only a rectangular border around the target in the image is given. In the case of sparse targets, this type of method can accurately give the target position information. However, in the case of dense targets, the detection positioning frame appears Regions overlap, and other target pixels will be included in the target bounding box, resulting in significantly lower target classification and localization accuracy

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
  • Image recognition method and device and storage medium
  • Image recognition method and device and storage medium
  • Image recognition method and device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0106] see figure 1 , a schematic diagram of an image recognition method provided in an embodiment of the present application, as shown in the figure, the method includes steps S101 to S104:

[0107] S101. Read an image to be processed, and perform preprocessing on the image to be processed to obtain a first image;

[0108] S102, load the weight coefficient of the deep neural network;

[0109] S103, input the first image into the deep neural network, perform prediction of the recognition target, and obtain a prediction result;

[0110] S104. Draw the recognition target in the image to be processed according to the prediction result.

[0111] It should be noted that the prediction content of the identified target includes one or a combination of the following: target category, position of the target rectangular frame or position information of the target pixel level.

[0112] In the method of this embodiment, the image to be processed is firstly preprocessed, and the preproc...

Embodiment 2

[0232] Based on the same inventive concept, the embodiment of the present invention also provides an image recognition device, such as Figure 10 As shown, the device includes:

[0233] The data set construction module 1001 is used to collect training images, generate annotation information of the training images, and the annotation information includes target position information labeling, target mask labeling, and target annotation files;

[0234] The algorithm training module 1002 is used to generate the weight coefficient of the deep neural network according to the image output by the data construction module;

[0235] The algorithm testing module 1003 is configured to predict the target in the image to be processed according to the weight coefficient.

[0236] It should be noted that the device provided in Embodiment 2 and the method provided in Embodiment 1 belong to the same inventive concept, solve the same technical problem, and achieve the same technical effect. The...

Embodiment 3

[0256] Based on the same inventive concept, the embodiment of the present invention also provides an image recognition device, such as Figure 14 As shown, the device includes:

[0257] Including memory 1402, processor 1401 and user interface 1403;

[0258] The memory 1402 is used to store computer programs;

[0259] The user interface 1403 is configured to interact with the user;

[0260] The processor 1401 is configured to read the computer program in the memory 1402, and when the processor 1401 executes the computer program, implements:

[0261] reading the image to be processed, and performing preprocessing on the image to be processed to obtain a first image;

[0262] Load the weight coefficients of the deep neural network;

[0263] Inputting the first image into the deep neural network, performing prediction of the recognition target, and obtaining a prediction result;

[0264] Draw the recognition target in the image to be processed according to the prediction result...

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 an image recognition method and device and a storage medium, which are used for reducing the omission ratio and improving the accuracy of target detection. The image recognition method disclosed by the invention comprises the steps of reading a to-be-processed image, and preprocessing the to-be-processed image to obtain a first image; loading a weight coefficient of the deep neural network; inputting the first image into the deep neural network, and predicting an identification target to obtain a prediction result; and drawing an identification target in the to-be-processed image according to the prediction result. The invention further provides an image recognition device and a storage medium.

Description

technical field [0001] The present application relates to the field of information processing, and in particular to an image recognition method, device and storage medium. Background technique [0002] With the development of computer vision technology, target detection technology is more and more widely used in production and life. With the expansion of application fields, the requirements for target detection accuracy in engineering applications are getting higher and higher, especially in the aerospace field. Object detection methods cannot meet their high precision requirements. In the prior art, only a rectangular border around the target in the image is given. In the case of sparse targets, this type of method can accurately give the target position information. However, in the case of dense targets, the detection positioning frame appears Regions overlap, and other target pixels are included in the bounding box of the target, resulting in significantly lower target c...

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): G06K9/00G06K9/32G06K9/40G06K9/62G06N3/08
CPCG06N3/084G06V20/40G06V10/30G06V10/25G06F18/214
Inventor 薛晗庆潘红九陈政王晓天王斌李凯赵翔宇赵媛心窦小明陈超尹琼底亚峰雷净刘萍
Owner BEIJING LINJIN SPACE AIRCRAFT SYST ENG INST