Method for automatically reading reading number of pointer meter

An automatic reading and instrument reading technology, applied in the field of machine learning and computer vision, can solve problems such as visual fatigue, laborious work, and easy misjudgment

Active Publication Date: 2018-09-21
NANJING UNIV
View PDF2 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The structure of the pointer instrument is simple, easy to maintain, high in precision, and resistant to electromagnetic interference. It is widely used in today's test and control systems. Because it has no data interface function, it generally uses the method of manual reading, but this is undoubtedly It is a laborious work, the manual reading method is inefficient, and reading for a long time will cause visual fatigue, which is prone to misjudgment, so the automatic reading recognition of pointer instruments has been a long-term research problem
At present, many methods and methods cannot escape the strict requirements on the instrument image. First, the image is required to be high-definition, and there is no suitable lighting and little background interference. The pointer and the dial can be easily separated, and the scale can be obtained. However, it is difficult to achieve this condition in practical applications.
However, many current studies are based on well-prepared experimental data, and few studies on the recognition of pointer instrument readings in the real field

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 automatically reading reading number of pointer meter
  • Method for automatically reading reading number of pointer meter
  • Method for automatically reading reading number of pointer meter

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0078] This embodiment includes the following parts:

[0079] Step 1, image acquisition and preprocessing.

[0080] Load the collected image containing the front of the instrument as the original image, and convert the original image into a grayscale image according to the color value. Reducing the number of channels can significantly reduce the time-consuming detection, using the formula Convert an RGB three-channel image to a single-channel grayscale image. Because the size of the original image is too large, there will be many windows to be selected, which is not very helpful to the detection instrument window, so the size is reduced proportionally, and the original image is resized according to the original length and width relationship to row_cur*row_col=640* 480 or 480*640, conversion ratio col_pre and row_pre represent the column size and row size of the image before resizing.

[0081] Step 2, the training of the classifier.

[0082] Through the pre-collected imag...

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 method for automatically reading a reading number of a pointer meter. The method comprises the steps of performing image collection, shooting an image comprising a meter front surface by utilizing a camera device to serve as a training sample; performing sample making, making the collected image into an image of a meter part through manual calibration to serve as a positive sample, and randomly generating an image of a background part to serve as a negative sample; performing image feature extraction, extracting digital features from the positive and negative samplesto serve as a training set; performing classifier training, obtaining a trained classifier based on the training set, for a new input image, predicting the most possible position of the meter by utilizing the classifier SVM, and extracting a meter image; performing image preprocessing, removing interferences such as shadows and the like for the extracted meter image through an image processing method; identifying a pointer and a dial plate of the meter, and obtaining angle and position information of the pointer; and obtaining a reading number value by utilizing the information of the pointer.

Description

technical field [0001] The invention belongs to the technical field of machine learning and computer vision, and in particular relates to a method for automatically reading pointer meters. Background technique [0002] With the rapid development of machine learning and machine vision, it is becoming more and more important for computers to understand the content of images today. In 2016, image data accounted for 85% of the entire Internet. Relying on manpower to process the huge amount of information is undoubtedly a costly and low-return work. It is a reasonable processing method to allow the computer to recognize the image content by itself. In electrical equipment, the monitoring of instrument readings is very important, and if the instrument is in an abnormal state, it will have serious consequences. However, if only relying on personnel to inspect and take readings, it will be a problem in terms of efficiency and manpower consumption. If a method can be found that ca...

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): G06K9/34G06K9/62
CPCG06V30/153G06F18/214G06F18/2411
Inventor 杨育彬周杨浩李瑮刘一帆
Owner NANJING UNIV
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