Restoration of blurred images for surface roughness evaluation using machine vision

An attempt is made to evaluate the surface roughness of uniformly moving machined surface (grinding, milling) using machine vision technique. In the case of moving surfaces the images are likely to blur due to the relative motion between the CCD camera and the object to be captured. Hence the degraded image has to be restored by removing distortion due to motion before subsequent analysis. In this work, image blur due to motion is considered, in particular, blur that occurs when the motion is uniform at constant speed and in a fixed direction. The blurred image is modeled as a convolution between the original image and a known point spread function. The Richardson–Lucy Restoration algorithm, a method of estimation based on Bayes theorem has been used to correct the image. The algorithm is tested in simulations and in practical experiments. A simulation gives complete control over the setup and enables to test the performance of the algorithm. The quantification of roughness for restored images are performed using the statistical parameters such as spatial frequency, arithmetic average of gray level and standard deviation after pre-processing. An Artificial Neural Network (ANN) was used with these three statistical parameters as input to predict the vision roughness. Finally, vision roughness values calculated using the deblurred images are compared with the stylus roughness value. An analysis based on the comparison to understand the validity of the present approach of estimation of surface roughness based on the digitally processed images for implementation in practice, is presented in this paper.     

Surface roughness classification using image processing

Surface roughness is an important factor in determining the satisfactory functioning of the machined components. Conventionally the surface roughness measurement is done with a stylus instrument. Since this measurement process is intrusive and is of contact type, it is not suitable for online measurements. There is a growing need for a reliable, online and non-contact method for surface measurements. Over the last few years, advances in image processing techniques have provided a basis for developing image-based surface roughness measuring techniques. Based upon the vision system, novel methods used for human identification in biometrics are used in the present work for characterization of machined surfaces. The Euclidean and Hamming distances of the surface images are used for surface recognition. Using a CCD camera and polychromatic light source, low-incident-angle images of machined surfaces with different surface roughness values were captured. A signal vector was generated from image pixel intensity and was processed using MATLAB software. A database of reference images with known surface roughness values was established. The Euclidean and Hamming distances between any new test surface and the reference images in the database were used to predict the surface roughness of the test surface.     

Uncertainty analysis of roughness standard calibration using stylus instruments

A stylus instrument was characterized and calibrated, including a dynamic calibration of the probe. This stylus instrument was used to calibrate ten roughness standards for six surface roughness parameters. The sensitivity of each parameter of each standard to such measurement conditions as stylus geometry, measurement force, cut-off wavelength, and so forth was determined experimentally. These results were used for an uncertainty evaluation of each parameter for each roughness standard. It is shown that the manufacturers’ specification for the stylus instrument (2% uncertainty in roughness parameters) is approximately correct for the most commonly used samples and parameters, but the uncertainty may range from 0.03% (for sinusoidal profiles) to 100% (for very fine surfaces), depending upon the standard and parameter to be calibrated.     

A Study of Computer Vision for Measuring Surface Roughness in the Turning Process

The paper presents a system for measuring the surface roughness of turned parts using a computer vision system. The images of specimens grabbed by the computer vision system are processed to obtain parameters of their grey levels (spatial frequency, arithmetic mean value, and standard deviation). These parameters are used as input data to a polynomial network. Using the trained polynomial network, the experimental result shows that the surface roughness of a turned part made of S55C steel, measured by the computer vision system over a wide range of turning conditions, can be obtained with reasonable accuracy, compared to that measured by a traditional stylus method. Compared with the stylus method, the computer vision system constructed is a useful method for measuring the surface roughness of this material faster, at a lower cost, and with lower environmental noise.     

Evaluation of discrete displacement transducer by spectral analysis of surface roughness

A discrete displacement transducer using an optical technique and attached to a commonly used stylus instrument is described. The proposed transducer plays an important role in generating a digital output of surface profile from the viewpoint of sampling error. The working accuracy is discussed with reference to the spectral analysis of surface roughness.     

Noncontact roughness measurement of turned parts using machine vision

The surface roughness of turned parts is usually measured using the conventional stylus type instruments. These instruments, although widely accepted, have several limitations such as low speed measurement, contacting in nature, requiring vibration-free environment, etc. Machine vision methods of roughness measurement are being developed worldwide due to their inherent advantages, including noncontact measurement, high information content, rapid measurement, and surface measurement capability. In past research, area-based light scattering method and gray scale line intensity measurement have been developed for roughness assessment using machine vision. Such methods, however, produced redundant data when applied to measure roughness of turned parts. In this paper, an alternative method of roughness measurement using the 2-D profile extracted from an edge image of the workpiece surface is proposed. Comparison with a stylus type instrument shows a maximum difference of 10% in the measurement of average roughness R _a using the vision method.     

Evaluation of image based Abbott–Firestone curve parameters using machine vision for the characterization of cylinder liner surface topography

In this paper, a method is proposed for the evaluation of image based Abbott–Firestone curve parameters aiming to characterize the cylinder bore surface topography using machine vision. Plateau honing experiments are performed to generate sixteen cylinder liners with different surface topographies and the 2-D and 3-D Abbott–Firestone parameters are measured using a stylus instrument and Coherence Scanning Interferometer (CSI), respectively. The images are captured from the corresponding portions of the cylinder liner surfaces using a Charge Coupled Device (CCD) camera connected with different microscopic attachments. The captured images are filtered using a Butterworth high pass filter followed by the adaptation of the double step Gaussian filtering procedure specified by the ISO 13565-1. An Abbott–Firestone curve is constructed by finding the cumulative of the intensity histogram of the filtered images. Five image based parameters are evaluated from the constructed Abbott curve by adapting the procedures presented in ISO 13565-2. The computed image based Abbott–Firestone curve parameters are observed to bear a statistically significant correlation with the measured 2-D and 3-D Abbott–Firestone curve parameters. An artificial neural network (ANN) is trained and tested to arrive at the actual values of the Abbott–Firestone curve parameters using the computed image based feature parameters. The results indicate that the multiple surface topography parameters of the cylinder bore surface could be estimated/predicted with a reasonable accuracy using machine vision technique coupled with ANN.     

Laser optical roughness measurement

A method for non-contacting profile assessment and roughness measurements of engineering surfaces is presented. The well known effect of focusing a light beam on the surface is used, at the same time compensating for the effect on the measuring results of varying material colours by appropriately processing the measuring signal. The sensor and the transducer are connected with the data-processing unit of a stylus instrument, which provides for a simple procedure of recording the roughness profile and calculating standard roughness parameters from it. The measuring results thus obtained agree well with the results of traditional stylus instruments.     

Measurement of the surface roughness of wood based materials used in furniture manufacture

The objective of this work was to evaluate surface quality of wood based materials used to manufacture furniture units in Singapore. Various type commercially produced composite panels including particleboard, medium density fibreboard (MDF), plywood in addition to ten different solid wood species which are commonly used in furniture production were considered for the experiments. A stylus type profilometer and 3D image analyzer were employed to determine surface roughness of the samples. Medium density fibreboard (MDF) samples resulted in the smoothest surface with an across the sandmark average roughness (R_a) value of 5.07 μm, while corresponding value for plywood specimens was 8.09 μm among the composite panel samples. In the case of solid wood samples, measurements taken along and across the sandmark from the surface of the specimens measured by the stylus type profilometer, balau had the roughest surface with an R_a value of 9.85 μm across the sandmark followed by beech and walnut. Pine specimens along with ash, cherry and nyatoh resulted in relatively smooth surface values. Correlation between measurements taken by two different methods, namely stylus and 3D scanning showed a good agreement with each other. Based on the findings in this work it appears that both methods can be successfully used to evaluate and to get objective numerical values on surface quality of these samples so that such initial data can be used as quality control tool to have more effective further manufacturing steps in furniture production.     

Rapid measurement of surface roughness for face-milling aluminum using laser scattering and the Taguchi method

In this paper, an experimental model for the rapid measurement of surface roughness (R_rms) in CNC face-milling specimens using the laser speckle method and digital image processing is established. The specimens used in this study were made of 6061-aluminum alloys through the high-speed face-milling process. In order to evaluate the effect of machining conditions, such as the feed rate, the spindle speed, the depth of cut, and the material of the cutting tool on the roughness of the specimens, the Taguchi method was used to determine the optimal parameters for machining. The laser radiation results in the speckle structure formed in the space when coherent light is scattered through an optically rough surface. The features of the speckles depend on the characteristics of the rough surfaces. Hence, the experimental work for the roughness measurement is based on the speckle effect. The experimental setup in this study consisted of a He-Ne laser, a ground glass, a CCD camera, and a digital image processing system developed using the Virtual Basic language. Computer evaluation of the speckle images revealed the values of R_rms rapidly. This study proposed a precise and non-contact optical method for evaluating the surface roughness from 0.20 to 0.60 μm.     

Analysis of surface roughness parameters digital image identification

The article analyses some surface roughness parameters of metal parts determining the ability of the surface of digital image identification, covariance functions and Wavelet’s wave theory. Expressions of covariance functions are formed using random functions, made by spreading digital image pixel arrays by columns in the form of individual vectors. The digital images used for research may vary in scale, because the frequencies of colour waves with individual pixels remain constant in the images, therefore, the image change does not influence the scale in computing covariance functions. The colour spectrum of RGB format was applied to identify the surface images of the parts. There was analysed the influence of individual RGB colour tensor components on the estimates of digital image covariance functions. The identity of digital images was evaluated by the change of correlation coefficient values in the range of RGB colours. The software was applied to compute the above process.     

In-cycle monitoring of tool nose wear and surface roughness of turned parts using machine vision

Tool wear has been extensively studied in the past due to its effect on the surface quality of the finished product. Vision-based systems using a CCD camera are increasingly being used for measurement of tool wear due to their numerous advantages compared to indirect methods. Most research into tool wear monitoring using vision systems focusses on off-line measurement of wear. The effect of wear on surface roughness of the workpiece is also studied by measuring the roughness off-line using mechanical stylus methods. In this work, a vision system using a CCD camera and backlight was developed to measure the surface roughness of the turned part without removing it from the machine in-between cutting processes, i.e. in-cycle. An algorithm developed in previous work was used to automatically correct tool misalignment using the images and measure the nose wear area. The surface roughness of turned parts measured using the machine vision system was verified using the mechanical stylus method. The nose wear was measured for different feed rates and its effect on the surface roughness of the turned part was studied. The results showed that surface roughness initially decreased as the machining time of the tool increased due to increasing nose wear and then increased when notch wear occurred.     

Measurement of surface roughness on rough machined surfaces using spectral speckle correlation and image analysis

A speckle pattern is formed when a rough surface is illuminated with coherent light. The properties of this pattern can be used in the calculation of roughness parameters. Spectral speckle correlation (SSC) is a technique applicable to the measurement of roughness on rough machined surfaces. This paper presents the SSC obtained from measurements on specimens with a surface roughness in range R_a = 0.5–5 μm. The measurement results correspond to reference measurements made using a stylus instrument, Form Talysurf.     

SPECKLE METHOD EVALUATION OF THE ROUGHNESS OF ELECTRODISCHARGE MACHINED SURFACES

Abstract

This paper describes the applicability of the speckle method to evaluate the roughness of surfaces produced by non-traditional machining processes such as EDM, ECM or USM, where material removal occurs randomly. The method is based on the digital correlation of two speckle images produced by interference phenomena arising when a coherent light beam is incident on a rough surface. The basic principle of the speckle pattern correlation is presented and the theoretical analysis based on a hypothesis concerning the morphology of the machined surface is reported. An experiment has been set-up in order to assess the feasibility of applying the speckle method to evaluate the roughness of machined surfaces. In the experimental tests, two speckle patterns produced from the same rough surface under two different illumination conditions have been correlated. The two different conditions have been obtained by varying the angle of incidence of coherent light on the surface being analysed. The roughness of electrodischarge machined surfaces, as measured by the speckle method, is found to be in good agreement with that of a stylus instrument.     

Analysis of profile measurement techniques employed to surfaces planed by an active machining system

The importance of a reliable and robust surface profile measurement system in the inspection of surface finish is beyond any doubt. For years, visual inspection has been employed in industries to determine the quality of surface finish. Since, in most cases, it fails to ensure a consistent minimum standard of finish quality, mechanical stylus based measurement systems have successfully taken over from human inspection. However, in recent years, the trend is to explore other techniques for conducting surface profile measurements. Non-contact optical methods have emerged as one of the leading candidates. In this paper, capabilities of two optical profile measurement methods (namely, light-sectioning and two-image photometric stereo) have been explored for surfaces machined using an active machining system. These profile measurement results have been compared to the ones obtained from a conventional mechanical stylus instrument. An industry-standard Talysurf CLI system has been used to provide the benchmark, traceable to NPL standards, for the measurements. Suitability of different measurement techniques have been discussed based on the results obtained.     

Precision surface characterization for finish cylindrical milling with dynamic tool displacements model

In this work a new approach to surface roughness parameters estimation during finish cylindrical end milling is presented. The proposed model includes the influence of cutting parameters, the tool’s static run out and dynamic phenomena related to instantaneous tool deflections. The modeling procedure consists of two parts. In the first stage, tool working part instantaneous displacements are estimated using an analytical model which considers tool dynamic deflections and static errors of the machine – tool-holder – tool system. The obtained height of the tool’s displacement envelope is then applied in the second stage to the calculation of surface roughness parameters. These calculations assume that in the cylindrical milling process, two different mechanisms of surface profile formation exist. Which mechanism is present is dependent on the feed per tooth and the maximum height of the tool’s displacement envelope. The developed model is validated during cylindrical milling of hardened hot-work tool steel 55NiCrMoV6 using a stylus profiler and scanning laser vibrometer over a range of cutting parameters. The surface roughness values predicted by the developed model are in good agreement with measured values. It is found that the employment of a model which includes only the effect of static displacements gives an inferior estimation of surface roughness compared to the model incorporating dynamic tool deflections.     

Measurement and evaluation of surface roughness based on optic system using image processing and artificial neural network

The traditional devices, used to measure the surface roughness, are very sensitive, and they are obtained by scratching the surface of materials. Therefore, the optic systems are used as alternatives to these devices to avoid the unwanted processes that damage the surface. In this study, face milling process was applied to American Iron and Steel Institute (AISI) 1040 carbon steel and aluminium alloy 5083 materials using the different tools, cutting speeds and depth of cuts. After these processes, surface roughness values were obtained by the surface roughness tester, and the machined surface images were taken using a polarise microscope. The obtained images were converted into binary images, and the images were used as input data to train network using the MATLAB neural network toolbox. For the training networks, log-sigmoid function was selected as transfer function, scaled conjugate gradient (SCG) algorithm was used as training algorithm, and performance of the trained networks was achieved as an average of 99.926 % for aluminium alloy (AA) 5083 aluminium and as an average of 99.932 % for AISI 1040 steel. At the end of the study, a prediction programme for optical surface roughness values using MATLAB m-file and GUI programming was developed. Then, the prediction programme and neural network performance were tested by the trial experiments. After the trial experiments, surface roughness values obtained with stylus technique for the carbon steel and aluminium alloy materials were compared with the developed programme values. When the developed programme values were compared with the experimental results, the results were confirmed each other at a rate of 99.999 %.     

Form Error Characterisation by an Optical Profiler

Form errors are deviations of the machined surface from the geometrical surface excluding position errors, waviness and roughness. From a functional point of view, as for surface roughness, form error characterisation is also important. In the present work, an optical profiler is used to measure and numerically characterise form errors such as roundness and cylindricity of cylindrical surfaces. A double orientation method using mean value analysis has been applied to separate the workpiece error from the spindle error during roundness measurement. Software is developed for data generation, fitting the reference data for assessing form errors in terms of statistical and functional parameters including new parameters. An optical profiler measures all the surface irregularities and hence can be used to study both micro and macro errors of the profile measured. A study of both roughness and roundness parameters along the circumferential direction is made for the unfiltered signal using different filter cut-off values. It is known that filtering greatly affects the value of the form error parameters measured. The form measurements obtained by the optical profiler are compared with the stylus profiler and the results are presented.     

Noncontact surface roughness measurement of engineering surfaces by total integrated infrared scattering

As precision engineering surfaces are gaining in importance in industry, so are the surface quality requirements. These surfaces have rms roughness typically ranging from some nanometers up to a few micrometers. Although numerous techniques exist for rough surface characterization, from traditional line-scanning stylus profilometers to modern three-dimensional (3-D) measurement instruments, there is a need for a fast, area-covering technique. An efficient method for the characterization of smooth surfaces is elastic light scattering. At visible wavelengths, the limits on roughness range and spatial frequency range make the method unsuitable for characterizing engineering surfaces. By increasing the wavelength of the incident light from the visible to the infrared, elastic light scattering turns out to be applicable for engineering surfaces. We have used total integrated scattering at 10.6 μm wavelength to measure rms roughness up to two micrometers. In this paper, the instrument design and properties are reviewed. We also present results from measurements on ground steel surfaces. Excellent correspondence with mechanical stylus measurements exists for surfaces with rms roughness in the range from 0.1–1.7 μm. The technique shows potential for rapid quality inspection of engineering surfaces.     

Determination of Surface Roughness in Wire and Arc Additive Manufacturing Based on Laser Vision Sensing

Wire and arc additive manufacturing (WAAM) shows a great promise for fabricating fully dense metal parts by means of melting materials in layers using a welding heat source. However, due to a large layer height produced in WAAM, an unsatisfactory surface roughness of parts processed by this technology has been a key issue. A methodology based on laser vision sensing is proposed to quantitatively calculate the surface roughness of parts deposited by WAAM. Calibrations for a camera and a laser plane of the optical system are presented. The reconstruction precision of the laser vision system is verified by a standard workpiece. Additionally, this determination approach is utilized to calculate the surface roughness of a multi-layer single-pass thin-walled part. The results indicate that the optical measurement approach based on the laser vision sensing is a simple and effective way to characterize the surface roughness of parts deposited by WAAM. The maximum absolute error is less than 0.15 mm. The proposed research provides the foundation for surface roughness optimization with different process parameters.