Laser Digitizer/Stereo Vision Methods for Simultaneous Measurement/Analysis of Sheet Metal Forming Strain/Geometry

2005 ◽  
Author(s):  
H.-L. Chan ◽  
J. P. Mitchell ◽  
A. D. Spence ◽  
M. P. Sklad ◽  
D. W. Capson
Keyword(s):  
Computer Vision ◽  
Simultaneous Measurement ◽  
Metal Forming ◽  
Computer Processing ◽  
Practical Application ◽  
Welded Steel ◽  
Measurement Analysis ◽  
Touch Trigger Probe ◽  
Small Test ◽  
Laser Digitizer

Conventionally, strain has been measured using scribed, etched, or printed grids together with a ruler and dividers or a graduated, transparent Mylar® overlay. Touch trigger probe equipped Coordinate Measuring Machines (CMMs) are used to measure geometry. More recently, stereo computer vision has been proposed, but, because accuracy quickly diminishes with camera distance to the part, application has been limited to small test samples. On CMMs, laser digitizers are growing in use. To date, however, accurate, simultaneous measurement / analysis of both strain and geometry has not been reported. This paper presents two new CMM based research methods that address this. The laser digitizer method uses a circle grid, and the stereo computer vision approach uses a square grid. Parallel computing is used to accelerate computer processing time. A dual thickness tailor welded steel test dome, an aluminum test dome, and a wrinkled double layer truck part illustrate practical application.

scholarly journals Simultaneous measurement of flexible robotic link deflection and rotation using position sensing detectors

2021 ◽  
Author(s):  
George Balyasin
Keyword(s):  
Computer Vision ◽  
Simultaneous Measurement ◽  
Laser Diodes ◽  
Strain Gauges ◽  
Trigonometric Functions ◽  
New Materials ◽  
Positioning Accuracy ◽  
End Effector ◽  
Position Sensing ◽  
Optoelectronic Method

It is well known that the flexibility of robotic links causes deformation and reduced positioning accuracy of the robot at the end-effector. This problem becomes more complicated when the links are made of new materials, such as composite. Various approaches including strain gauges, fiber Bragg grating, computer vision and optoelectronics have been applied to solve the problem. This work proposes an optoelectronic method with a reduced number of position sensing detectors and laser diodes to determine both bending and torsion deformations of a robotic link. The attachment of two optoelectronic sensors on the link and the two laser dots produced on these sensors by diodes provide the data required for computations. The tip deflection values are obtained using trigonometric functions. Both iterative and explicit methods of link tip movement have been analysed for efficiency and accuracy. The methods have been validated experimentally and error analysis has been performed.

Soft-Sensing Model of Deformation of Welded Steel Structure Based on FLS-SVM and its Application

2014 ◽  
Vol 628 ◽  
pp. 152-156
Author(s):  
Ji Ping Lei ◽  
Jian Mei Chen
Keyword(s):  
Relative Error ◽  
Welding Current ◽  
Steel Structure ◽  
Support Vector ◽  
Adaptive Genetic Algorithm ◽  
Practical Application ◽  
Soft Sensing ◽  
Welded Steel ◽  
Welded Structure ◽  
Sensing Model

To effectively achieve rapid and high-precision measurements of the deformation of steel welded structure, multiple sets of the actual experimental data of steel welded structure are used as the samples, the soft-sensing model of deformation of welded steel structure, which uses the welding current I, the welding voltage U, the welding speed v and the flow of gas qm as arguments, is established by fuzzy least squares support vector machine, and adaptive genetic algorithm is used to optimize the number of positive gasification rules c and the parameters of kernel function σ, training, testing and practical application results show, the optimization of 200 steps, the training relative error which become saturated is 2.43%, the testing relative error is less than 2.45%.

WED 112 Seeing invisible parkinsonian tremor with a smartphone camera

2018 ◽  
Vol 89 (10) ◽  
pp. A12.1-A12
Author(s):  
O’Gorman Paschal ◽  
Williams Stefan ◽  
Fang Hui ◽  
Qahwaji Rami ◽  
Patel Parisa ◽  
...  
Keyword(s):  
Computer Vision ◽  
Optical Flow ◽  
Small Amplitude ◽  
Computer Processing ◽  
Advanced Technology ◽  
Computing Technique ◽  
Parkinsonian Tremor ◽  
Video Recordings ◽  
Third Law ◽  
Computer Processor

Arthur C Clarke’s ‘third law’ states that any sufficiently advanced technology is indistinguishable from magic. Computer vision is the processing of images or video by computer to extract useful information. A technique termed ‘Eulerian magnification’ involves amplification of tiny movements from video recordings, so that very small motions can become visible to the human eye. This has the potential to detect tremor that is of such small amplitude it cannot otherwise be seen. Crucially, the only hardware required is a camera and computer processor, items that are ubiquitous. There is only one previous report of Eulerian magnification applied to a simple video of a Parkinson’s patient, but Parkinsonian signs could clearly be seen in the pre-processing video, and no control video was shown. We present remarkable video in which no tremor is seen in either patient or control before processing, and yet a Parkinsonian tremor is revealed in patient but not control after amplification. Blinded clinician ratings detect a greater number of Parkinsonian tremors after computer processing. Furthermore, we report a method using an ‘optical flow’ computing technique that records pixel motion vectors, and enables the computer to measure the direction and relative amplitude of this amplified movement.

scholarly journals Small Network for Lightweight Task in Computer Vision: A Pruning Method Based on Feature Representation

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Yisu Ge ◽  
Shufang Lu ◽  
Fei Gao
Keyword(s):  
Neural Networks ◽  
Computer Vision ◽  
Feature Representation ◽  
Fine Tuning ◽  
Practical Application ◽  
Pruning Algorithm ◽  
Network Pruning ◽  
Pruning Strategy ◽  
Speed Up ◽  
Network Compression

Many current convolutional neural networks are hard to meet the practical application requirement because of the enormous network parameters. For accelerating the inference speed of networks, more and more attention has been paid to network compression. Network pruning is one of the most efficient and simplest ways to compress and speed up the networks. In this paper, a pruning algorithm for the lightweight task is proposed, and a pruning strategy based on feature representation is investigated. Different from other pruning approaches, the proposed strategy is guided by the practical task and eliminates the irrelevant filters in the network. After pruning, the network is compacted to a smaller size and is easy to recover accuracy with fine-tuning. The performance of the proposed pruning algorithm is validated on the acknowledged image datasets, and the experimental results prove that the proposed algorithm is more suitable to prune the irrelevant filters for the fine-tuning dataset.

scholarly journals Application of computer vision art based on emoticons in the field of animation

2021 ◽  
Vol 292 ◽  
pp. 03050
Author(s):  
Yang Cao ◽  
Yuxin Wang
Keyword(s):  
Computer Vision ◽  
Computer Animation ◽  
Practical Application ◽  
Integration Of Technology

Starting from the overview, this article explains the connotation of computer vision art and computer animation, and discusses the practical application of computer vision art in emoticon animation, in order to reflect the charm of the integration of technology and art, provide some value for the research in related fields.

scholarly journals Fixed Transmission Charges Based on the Degree of Network Utilization

Energies ◽  
2021 ◽  
Vol 14 (3) ◽  
pp. 614
Author(s):  
Roman Korab ◽  
Henryk Kocot ◽  
Henryk Majchrzak
Keyword(s):  
Power Flow ◽  
Superposition Principle ◽  
Test System ◽  
Practical Application ◽  
Postage Stamp ◽  
Network Utilization ◽  
Active Power Flow ◽  
Small Test ◽  
Power Flow Tracing ◽  
The Individual

The core objective of transmission tariffs is the recovery of costs related to the transport of electricity. A usual component of a tariff is a fixed charge that covers the costs of the network infrastructure. As many customers use the power grid, the rate of this charge should reflect, as closely as possible, the actual costs of supplying energy to the individual consumers. These costs result from which network elements have been used in delivering the electricity, and to what extent these elements have been used. Therefore, the fixed transmission rates should depend on the degree of network utilization. This article investigates definitions of the degree of network utilization based on the active power flow. To calculate the degree of network utilization, the flow of electricity on a branch must be decomposed into the streams flowing to individual customers. For this decomposition, two methods are examined: a power flow tracing method, based on the proportional sharing principle, and an incremental power flow method, based on the superposition principle. The analyzed methodology is applied to a small test system for conceptual discussions, as well as to the transmission network of the Polish power system, as an example of practical application. The results of this study were then compared with the commonly used “postage stamp” method. Finally, several practical aspects related to the potential implementation of the presented methodology are discussed.

scholarly journals A Modularized Architecture of Multi-Branch Convolutional Neural Network for Image Captioning

Electronics ◽  
2019 ◽  
Vol 8 (12) ◽  
pp. 1417 ◽  
Author(s):  
Shan He ◽  
Yuanyao Lu
Keyword(s):  
Neural Network ◽  
Computer Vision ◽  
Convolutional Neural Network ◽  
Language Processing ◽  
State Of The Art ◽  
Input Image ◽  
Simple Design ◽  
Complete Conversion ◽  
Practical Application ◽  
Image Captioning

Image captioning is a comprehensive task in computer vision (CV) and natural language processing (NLP). It can complete conversion from image to text, that is, the algorithm automatically generates corresponding descriptive text according to the input image. In this paper, we present an end-to-end model that takes deep convolutional neural network (CNN) as the encoder and recurrent neural network (RNN) as the decoder. In order to get better image captioning extraction, we propose a highly modularized multi-branch CNN, which could increase accuracy while maintaining the number of hyper-parameters unchanged. This strategy provides a simply designed network consists of parallel sub-modules of the same structure. While traditional CNN goes deeper and wider to increase accuracy, our proposed method is more effective with a simple design, which is easier to optimize for practical application. Experiments are conducted on Flickr8k, Flickr30k and MSCOCO entities. Results demonstrate that our method achieves state of the art performances in terms of caption quality.

scholarly journals Angiographic film substraction using a laser digitizer and computer processing

1998 ◽  
Vol 11 (4) ◽  
pp. 159-167 ◽  
Author(s):  
John M. Boone ◽  
Neva M. Corrigan ◽  
Stephen T. Hecht ◽  
Daniel P. Link
Keyword(s):  
Computer Processing ◽  
Laser Digitizer

scholarly journals Direct Metal Forming of a Microdome Structure with a Glassy Carbon Mold for Enhanced Boiling Heat Transfer

Micromachines ◽  
2018 ◽  
Vol 9 (8) ◽  
pp. 376 ◽  
Author(s):  
Jun Kim ◽  
Dongin Hong ◽  
Mohsin Badshah ◽  
Xun Lu ◽  
Young Kim ◽  
...  
Keyword(s):  
Heat Transfer ◽  
Glassy Carbon ◽  
Metal Forming ◽  
Boiling Heat Transfer ◽  
Applied Pressure ◽  
Practical Application ◽  
Forming Temperature ◽  
Master Pattern ◽  
Enhanced Boiling ◽  
Thermal Reflow

The application of microtechnology to traditional mechanical industries is limited owing to the lack of suitable micropatterning technology for durable materials including metal. In this research, a glassy carbon (GC) micromold was applied for the direct metal forming (DMF) of a microstructure on an aluminum (Al) substrate. The GC mold with microdome cavities was prepared by carbonization of a furan precursor, which was replicated from the thermal reflow photoresist master pattern. A microdome array with a diameter of 8.4 μm, a height of ~0.74 μm, and a pitch of 9.9 μm was successfully fabricated on an Al substrate by using DMF at a forming temperature of 645 °C and an applied pressure of 2 MPa. As a practical application of the proposed DMF process, the enhanced boiling heat transfer characteristics of the DMF microdome Al substrate were analyzed. The DMF microdome Al substrate showed 20.4 ± 2.6% higher critical heat flux and 34.1 ± 5.3% higher heat transfer coefficient than those of a bare Al substrate.

scholarly journals Simultaneous measurement of flexible robotic link deflection and rotation using position sensing detectors

2021 ◽  
Author(s):  
George Balyasin
Keyword(s):  
Computer Vision ◽  
Simultaneous Measurement ◽  
Laser Diodes ◽  
Strain Gauges ◽  
Trigonometric Functions ◽  
New Materials ◽  
Positioning Accuracy ◽  
End Effector ◽  
Position Sensing ◽  
Optoelectronic Method

It is well known that the flexibility of robotic links causes deformation and reduced positioning accuracy of the robot at the end-effector. This problem becomes more complicated when the links are made of new materials, such as composite. Various approaches including strain gauges, fiber Bragg grating, computer vision and optoelectronics have been applied to solve the problem. This work proposes an optoelectronic method with a reduced number of position sensing detectors and laser diodes to determine both bending and torsion deformations of a robotic link. The attachment of two optoelectronic sensors on the link and the two laser dots produced on these sensors by diodes provide the data required for computations. The tip deflection values are obtained using trigonometric functions. Both iterative and explicit methods of link tip movement have been analysed for efficiency and accuracy. The methods have been validated experimentally and error analysis has been performed.

Sign in / Sign up

Export Citation Format

Share Document