Search results

Researchers in the past have used Fourier transformation method to determine the in‐plane displacement components from moiré fringes generated by a pair of overlapping circular gratings. In this approach it is necessary to assume that the transmittance is sinusoidal. The purpose of this paper is to propose a graphical method for determining the 2D displacement components from the moiré patterns more easily instead of the complex Fourier transformation method.

The moiré patterns were spatially transformed from Cartesian‐to‐polar coordinate system. The morphological grayscale dilation operation was used to eliminate the residual gratings in the transformed pattern while preserving the moiré fringes. The center line of the moiré fringe was fitted with a sine curve and the in‐plane displacement values were determined directly from the peak‐to‐valley height and the position of the peak in the fitted curve.

Experimental results showed that the proposed moiré pattern analysis method is able to give in‐plane displacement accuracies of 0.002 mm in the x‐direction and 0.01 in the y‐direction without the need for complex computation.

Resolution of the proposed method is limited only by the resolution of the imaging system.

The proposed graphical method for determining 2D displacement components from the moiré patterns can be applied to low‐frequency circular gratings whose transmittance is not sinusoidal.

The graphical analysis method is novel and allows the displacements components to be determined more easily.

The nose radii of cutting inserts are normally measured using a profile projector or toolmaker's microscope. Since only a sector of a circle is available for the measurement using such instruments, the radii determined from these methods are inaccurate. The purpose of this paper is to present an alternative method of determining the nose radii more accurately using machine vision.

The 2D images of the cutting inserts were captured using a CCD camera with the aid of back lighting. The tool nose center in each digitized image was located based on the tool geometry. The curved nose profile was transformed into a linear profile using polar‐radius transformation. The nose radius was then varied within ten pixels of the nominal radius and the average deviation from a straight line profile in the nose region in the polar‐radius plot was evaluated. The radius corresponding to the minimum average deviation is identified as the most accurate radius value.

For the 15 simulated images of cutting inserts tested, the error in radii determine by the proposed method varied from −4.9 to 3.7 percent. But, the radii were about 9 to 22 percent higher than those measured using the profile projector on commercially available inserts. The radii measured using the profile projector was closer to the nominal radii with an average deviation of −3.2 percent compared to those measured using the proposed method.

The cutting inserts must be clean and free from dust particles when capturing the images; and the insert must be aligned accurately so that the plane of the nose profile is perpendicular to the optical axis of the CCD camera.

The proposed method can be used to determine the nose radii accurately. If the exact nose radius of an insert is known, the tool path can be programmed precisely to obtain high‐dimensional accuracy in the finished product.

The paper shows how a new method of determining the tool nose radii of cutting inserts more accurately compared to the conventional methods, based on a sector of the nose profile, has been developed.

This paper aims to propose a non-contact method using machine vision for measuring the surface roughness of a rotating workpiece at speeds of up to 4,000 rpm.

A commercial digital single-lens-reflex camera with high shutter speed and backlight was used to capture a silhouette of the rotating workpiece profile. The roughness profile was extracted at sub-pixel accuracy from the captured images using the moment invariant method of edge detection. The average (Ra), root-mean square (Rq) and peak-to-valley (Rt) roughness parameters were measured for ten different specimens at spindle speeds of up to 4,000 rpm. The roughness values measured using the proposed machine vision system were verified using the stylus profilometer.

The roughness values measured using the proposed method show high correlation (up to 0.997 for Ra) with those determined using the profilometer. The mean differences in Ra, Rq and Rt between the two methods were only 4.66, 3.29 and 3.70 per cent, respectively.

The proposed method has significant potential for application in the in-process roughness measurement and tool condition monitoring from workpiece profile signature during turning, thus, obviating the need to stop the machine.

The machine vision method combined with sub-pixel edge detection has not been applied to measure the roughness of a rotating workpiece.

Ghee, anhydrous milk fat, is chemically highly complex in nature. The authentication and characterization of edible fats and oils by routine chemical methods are highly laborious and time consuming. Fourier transform-mid-infrared (FT-MIR) spectroscopy has emerged as a predominant analytical tool in the study of edible fats/oils. However, sufficient attention has not been paid so far to spectral characterization of milk fat obtained from cow and buffalo milk. The purpose of this paper is to fill this void.

Ghee samples were prepared from cow and buffalo milk by the direct cream method. From each type of milk (cow and buffalo), 35 samples of ghee were prepared; thus, in total, 70 samples of ghee were obtained for the study. For assigning absorption bands in the IR spectrum, spectra of cow and buffalo ghee samples were acquired in the MIR region (4,000-650 cm⁻¹).

In FT-MIR spectra of ghee, 14 peaks were obtained at different positions and with varying intensities. They were at 3,005, 2,922, 2,853, 1,744, 1,466, 1,418, 1,377, 1,236, 1,161, 1,114, 1,098, 966, 870 and 721 cm⁻¹ for cow and buffalo ghee with almost equal intensity of absorption.

The finding of this study will be useful for characterization and authentication of ghee.

Application of IR spectral bands of ghee in the MIR region using a FT-infrared spectrometer to monitor the quality of ghee is suggested.

This paper describes how force‐guided robot can be implemented in the automated assembly of mobile phone. A case study was carried out to investigate the assembly operations and strategies involved. Force‐guided robot was developed and implemented in the real environment. Proportional‐based external force control with hybrid framework was developed and implemented to perform the compliant motion. In order to perform assembly operations, three basic force‐guided robotic skills are identified. These are stopping, alignment and sliding skills, where the motions are guided by the force feedback. The force‐guided robotic skills are combined and reprogrammed with fine motion planning to perform notch‐locked assembly. The system is optimized for high assembly speed while considering the constraints and limitations involved.

This paper describes the development of a force‐guided robot that uses the information of contact force to overcome component misalignment in the automated assembly of a mobile phone. Several possibilities of misalignment in the assembly of the back chassis and front housing of the mobile phone are studied. An assembly approach using force‐guided motion is implemented to perform the assembly task. The assembly operation was carried out in the presence of translational and rotational misalignment of the mating components. Experimental results show that the proposed assembly approach successfully performs the assembly task.

The purpose of this paper is to design a kernel convolution processor. High-speed image processing is a challenging task for real-time applications such as product quality control of manufacturing lines. Smart image sensors use an array of in-pixel processors to facilitate high-speed real-time image processing. These sensors are usually used to perform the initial low-level bulk image filtering and enhancement.

In this paper, using pulse-width modulated signals and regular nearest neighbor interconnections, a convolution image processor is presented. The presented processor is not only capable of processing arbitrary size kernels but also the kernel coefficients can be any arbitrary positive or negative floating number.

The performance of the proposed architecture is evaluated on a Xilinx Virtex-7 field programmable gate array platform. The peak signal-to-noise ratio metric is used to measure the computation error for different images, filters and illuminations. Finally, the power consumption of the circuit in different operating conditions is presented.

The presented processor array can be used for high-speed kernel convolution image processing tasks including arbitrary size edge detection and sharpening functions, which require negative and fractional kernel values.

The advancement of additive manufacturing technologies has resulted in producing parts of high quality and reduced manufacturing time. This paper aims to achieve a simultaneous optimal solution for build time and surface roughness as the output data and also to find the best values for the input data consisting of build orientation, extrusion width, layer thickness, infill percentage and raster angle.

For this purpose, the effects of process parameters on the response variables were investigated by the design of experiments approach to develop empirical models using response surface methodology. The experimental parts of this research were conducted using an inexpensive and locally assembled fused filament fabrication (FFF) machine. A total of 50 runs for 4 different geometries, namely, cylinder, prism, 3DBenchy and twist gear vase, were performed using the rotatable central composite design, and each process parameters were investigated in two levels to develop empirical models. Also, a novel optimization method, namely, the posterior-based method, was accomplished to find the best values for the response variables.

The results demonstrated that not only the build orientation and layer thickness have notable effects on both response variables but also build time is dependent on extrusion width and infill percentage. Low infill percentage and high extrusion width resulted in increasing build time. By reducing layer thickness and infill percentage while increasing extrusion width, parts of high-quality surface finish and reduced built time were produced. Optimum process parameters were found to be of build direction of 0°, extrusion width of 0.61 mm, layer thickness of 0.22 mm, infill percentage of 20% and raster angle of 0°.

Through the developed empirical models and by minimizing build orientation and layer thickness, and also considerations for process parameters, parts of high-quality surface finish and reduced built time could be produced on FFF machines. To compensate for increased build time because of reduction in layer thickness, extrusion width and infill percentage must have their maximum and minimum value, respectively.

The purpose of this study is to propose a pulse width based, in-pixel, arbitrary size kernel convolution processor. When image sensors are used in machine vision tasks, large amount of data need to be transferred to the output and fed to a processor. Basic and low-level image processing functions such as kernel convolution is used extensively in the early stages of most machine vision tasks. These low-level functions are usually computationally extensive and if the computation is performed inside every pixel, the burden on the external processor will be greatly reduced.

In the proposed architecture, digital pulse width processing is used to perform kernel convolution on the image sensor data. With this approach, while the photocurrent fluctuations are expressed with changes in the pulse width of an output signal, the small processor incorporated in each pixel receives the output signal of the corresponding pixel and its neighbors and produces a binary coded output result for that specific pixel. The process is commenced in parallel among all pixels of the image sensor.

It is shown that using the proposed architecture, not only kernel convolution can be performed in the digital domain inside smart image sensors but also arbitrary kernel coefficients are obtainable simply by adjusting the sampling frequency at different phases of the processing.

Although in-pixel digital kernel convolution has been previously reported however with the presented approach no in-pixel analog to binary coded digital converter is required. Furthermore, arbitrary kernel coefficients and scaling can be deployed in the processing. The given architecture is a suitable choice for smart image sensors which are to be used in high-speed machine vision tasks.

One of the prime beneficiaries of the diploma mill merchants is the publishing industry. A very high propor‐tion of the degrees sold are the result of press advertising.