scholarly journals A Novel Method for Measuring the Length of Hot Large Forgings Based on Machine Vision System

2021 ◽  
Author(s):  
Bangguo Wang
Keyword(s):  
Machine Vision ◽  
Measurement System ◽  
Vision System ◽  
Hot Forging ◽  
Nuclear Industry ◽  
Length Measurement ◽  
Three Dimension ◽  
Machine Vision System ◽  
Strip Edge ◽  
Novel Method

Abstract Large forgings can be applied to manufacture key parts in large equipment used in petrochemical, shipbuilding, aerospace, nuclear industry. At present, the size of hot large forgings is mainly measured by hand-hold calipers or mechanical gauges through contact measurement method. In order to realize non-contact length measurement of hot large forgings, a novel method for measuring the length of hot large forgings based on machine vision system is proposed. Firstly, the light strips are recognized according to the continuous characteristics of the light stripes in the image acquired by the hot forging dimension measurement system based on machine vision. Secondly, using the sub-pixel edge of light strips acquired by the improved sub-pixel edge detection algorithm, the three-dimension (3D) points of each edge of light strips are calculated, respectively. Lastly, the two-dimensional (2D) points projected from the 3D points onto the fitted plane are used for curve fitting, and the edge of hot part corresponding to each light strip edge is calculated according to the curvature of the fitted curves. The distance between the start point and the edge point on each light strip edge is considered as the length of hot forging at the corresponding edge of each light strip. The length measurement experiment shows that the method can be used to calculate the length of hot large forgings. The measurement error of the length measurement system when measuring part at room temperature is 0.547%. The time for measuring the length of hot forging is 6.8s.

A novel method based on machine vision system and deep learning to detect fraud in turmeric powder

2021 ◽  
pp. 104728
Author(s):  
Ahmad Jahanbakhshi ◽  
Yousef Abbaspour-Gilandeh ◽  
Kobra Heidarbeigi ◽  
Mohammad Momeny
Keyword(s):  
Deep Learning ◽  
Machine Vision ◽  
Vision System ◽  
Machine Vision System ◽  
Novel Method

2-D Circle Measurement System for Small Rule Parts Based on Machine Vision

2011 ◽  
Vol 291-294 ◽  
pp. 2624-2629 ◽  
Author(s):  
Qing Hua Wu ◽  
Na Dai ◽  
Tao He
Keyword(s):  
Machine Vision ◽  
Measurement System ◽  
Vision System ◽  
Measured Data ◽  
Machine Vision System ◽  
Mechanical Parts ◽  
Vision Measurement ◽  
Ring Radius ◽  
On Line ◽  
Calibration Mark

A Circle-shape is an important figure in most small rule mechanical parts, and usually be measured to get the radius or used as a stand calibration mark. In this paper, a 2-D circle measurement system for small rule mechanical parts based on machine vision is designed and built. The basic components and work principle of the machine vision measurement system are introduced, and the measurement produce is designed and discussed. An available algorithm for circle contour detected and fitted is described. Using this algorithm, the measurement software flow and architecture are built and the software system realized in the Microsoft visual studio program platform. Certainly, the calibration of machine vision system is introduced also. Using the system and method introduced above, an experiment is designed to measure the outer ring radius of one certain model bearing. The measured data is processed and analyzed. Through the experiment and result, it can be found that the measurement system can get relatively high precision and the measurement method is relatively steady, and the system precision and speed can be suit for the demand of on-line and real-time circle measurement.

Multifunctional stereoscopic machine vision system with multiple 3D outputs

2020 ◽  
Keyword(s):  
Machine Vision ◽  
International Symposium ◽  
Fast Track ◽  
Vision System ◽  
Machine Vision System ◽  
Stereoscopic Displays ◽  
Electronic Imaging

Fast track article for IS&T International Symposium on Electronic Imaging 2020: Stereoscopic Displays and Applications proceedings.

scholarly journals A real-time machine-vision system for monitoring the textile raising process

2005 ◽  
Vol 56 (8-9) ◽  
pp. 831-842 ◽  
Author(s):  
Monica Carfagni ◽  
Rocco Furferi ◽  
Lapo Governi
Keyword(s):  
Machine Vision ◽  
Real Time ◽  
Vision System ◽  
Machine Vision System ◽  
Time Machine

Application of Machine Vision System in the Precision Inspection

2012 ◽  
Vol 546-547 ◽  
pp. 1382-1386
Author(s):  
Yin Xia Liu ◽  
Ping Zhou
Keyword(s):  
Machine Vision ◽  
Vision System ◽  
Machine Vision System ◽  
Technical Problems ◽  
Application Program ◽  
Precision Inspection ◽  
Machine Parts

In order to promote the application and development of machine vision, The paper introduces the components of a machine vision system、common lighting technique and machine vision process. And the key technical problems are also briefly discussed in the application. A reference idea for application program of testing the quality of the machine parts is offered.

A machine vision system for on-line removal of contaminants in wool

Mechatronics ◽  
2006 ◽  
Vol 16 (5) ◽  
pp. 243-247 ◽  
Author(s):  
Zhenwei Su ◽  
Gui Yun Tian ◽  
Chunhua Gao
Keyword(s):  
Machine Vision ◽  
Vision System ◽  
Machine Vision System ◽  
On Line ◽  
Removal Of Contaminants

Research on Machine Vision in Forestry Pluck System

2010 ◽  
Vol 139-141 ◽  
pp. 2199-2202
Author(s):  
Xin Li ◽  
Chun Liang Zhang ◽  
Li Jun Li ◽  
Zhi Hu
Keyword(s):  
Machine Vision ◽  
Vision System ◽  
Area Ratio ◽  
Self Organization ◽  
Machine Vision System ◽  
Camellia Oleifera ◽  
Threshold Method ◽  
Input Feature ◽  
Ga Algorithm ◽  
Forestry Industry

Forestry industry is an important part of nation's economy. In this paper, a machine vision system is presented as a key module of Camellia oleifera pluck robot. In order to cut fruit image up from complicate background, SOFM neural network and gray thresh is used in image segmentation. In SOFM method, take R-B,G-R,G-B and hue H tunnel as input feature vectors, use self-organization network to clustering can get the best effect. in gray threshold method can take various of method to get the best threshold, such as PSO and GA algorithm, and MATLAB includes the toolboxes. At last use noise ratio, area ratio, divided time, Fourier boundary descriptors and other indicators to assess the accuracy of segmentation. The methods have the significance to the current and subsequent research of forestry pluck device.

Sign in / Sign up

Export Citation Format

Share Document