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

A rotor winding image detection method based on a binary image and model transfer learning

A binary image, transfer learning technology, applied in image enhancement, image analysis, image data processing and other directions, can solve the problems of insufficient training samples and the influence of rotor winding image detection accuracy, so as to reduce interference factors and improve detection accuracy. , the effect of strong accuracy

Inactive Publication Date: 2019-04-05
SOUTHEAST UNIV
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: The purpose of the invention is to solve the problem that the detection accuracy of the existing rotor winding image is affected by background and illumination factors, and there are insufficient training samples

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
  • A rotor winding image detection method based on a binary image and model transfer learning
  • A rotor winding image detection method based on a binary image and model transfer learning
  • A rotor winding image detection method based on a binary image and model transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] A rotor winding image detection method based on binary images and model transfer learning, as shown in Figure 1, specifically includes the following steps:

[0031] S1: Binarize the image of the rotor winding training set according to the RGB features, and perform de-interference processing on the binarized image to obtain the training set composed of the rotor winding binary image;

[0032] S2: Use the ImageNet dataset to pre-train the Inception-V3 model, and save the model to obtain the pre-trained model;

[0033] S3: Use the training set composed of the rotor winding binary image to fine-tune the fully connected layer of the pre-trained model to obtain the final model;

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 a rotor winding image detection method based on a binary image and model transfer learning, and the method comprises the steps: carrying out the binarization operation of a rotor winding training set image according to the RGB characteristics, carrying out the de-interference processing of the binary image, and obtaining a training set formed by a rotor winding binary image; using an ImageNet data set to pre-train an Inception-V3 model to obtain a pre-training model; Performing fine adjustment on a full connection layer of the pre-training model by using a training setformed by the rotor winding binary image to obtain a final model; Detecting the to-be-detected image subjected to binarization according to the RGB features by using the final model; Compared with theprior art, the method has the advantages that the influence of background and illumination factors on the detection precision is effectively avoided by integrating the two methods of binarized imageand transfer learning, the problem of insufficient training samples is solved, the detection time is shortened, and the detection accuracy is improved.

Description

technical field [0001] The invention relates to a digital image processing and visual detection method, and provides a detection method for the winding shape of a rotor hook, specifically, a detection method of a rotor winding image based on binary images and model transfer learning. Background technique [0002] With the continuous improvement of my country's industrial automation and intelligence level, motors, as key products that can convert electrical energy into mechanical energy, are widely used in large-scale electromechanical equipment in industries, agriculture, etc. In the equipment, the demand for the rotor, which is the core component of the motor rotation system, is also increasing. [0003] During the production process of the rotor, the hooks need to be surrounded by copper wires. Due to the irregular operation and the limitation of the properties of the copper wires, problems such as broken wires and missing hooks are prone to occur at the hooks, which will a...

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): G06T7/00G06T7/12G06N3/04
CPCG06T7/0004G06T7/12G06T2207/20081G06N3/045
Inventor 陈刚贾友彬张小国
Owner SOUTHEAST UNIV