Line-scan CCD Camera Research Articles

색채선별기 곡물 이미지 가시화 및 선별기법에 관한 연구

The color sorting technique utilizing the image processing method is very applicable tool to analyze motion of a free-falling object in many agricultural and industrial research fields. In the present study, we have developed an image processing system and algorithm to sort good quality rice grains effectively from the bad ones. The system employs a high speed rate line-scan CCD camera with 2K-pixels and worked with a high speed DSP and FPGA in-line. It can accumulate acquired line-scan image data and visualize each grain image clearly. As a result, we can easily calculate the number of pixels occupied by grain(=grain size), gray level and its correct position by visualizing grain images rapidly.

Surface profile measurement of moving objects by using an improved π phase-shifting Fourier transform profilometry

The π phase-shifting Fourier transform technique is introduced into the surface profile measurement of moving objects. A digital grating comprising two regions, which have a π phase shifting is projected onto the object. Two line-scan CCD cameras are used to capture two deformed fringe patterns with π phase shifting at the same time. As the object is moving, each point at the object surface can be captured twice. The digital correlation method is used to calibrate the experimental system. The zero-order component can be eliminated by subtracting intensities of the same surface point in two captured images. And then the phase can be extracted by Fourier transform without the disturbance of zero-order component. Experimental results demonstrate that this method is feasible for the moving surface profile detection and the measurable slope of height variation can be extended.

High-resolution line-scanning optical coherence microscopy

An optical coherence microscopy system based on line illumination and detection is demonstrated. The system uses a Linnik-type interferometer illuminated by a broadband Ti:sapphire laser and detected by a high-speed, line-scan CCD camera. This approach is less sensitive to incoherent scattering and sample motion than full-field imaging. Spatial resolutions of approximately 2 microm x approximately 3 microm(transverse x axial) are achieved. The sensitivity of the system is 93 dB with averaging over 30 line scans. En face real time, cellular-level imaging of biological tissues is demonstrated at approximately 2 frames/s.

Investigation of a measurement technique to estimate concentration and size of inclusions in droplets

Inhomogeneous droplets including small spherical inclusions are characterized by the estimation of droplet diameter, inclusion concentration and inclusion diameter and a measure for the polydispersity of the inclusions. In most cases, it is reasonable to assume that the material parameters are known and therefore the index of refraction and the shape of inclusions. In this paper, the results from a measurement technique are investigated. The method will be evaluated on the basis of light scattering measurements for a range of scattering angles. These measurements have been taken with a fast CCD line scan camera and appropriate optics. An attempt is made to derive information from these measurements only. The continuous wavelet transform, speckle image analysis and turbidity measurement methods are used to estimate the concentration and the diameter of the monodisperse polystyrene particles within a droplet. The droplets, generated by a drop-on-demand droplet generator, are nearly monodisperse. The volume concentration of the inclusions within the suspensions varies between 0.01% and 9%. The inclusions are monodisperse. However, it seems to be possible that they coagulate due to the fast fluid flows at the droplet generation. As a result, the technique may be used only for the estimation of average values of size and concentration of inclusions from the measurements.

Multispectral polarimetry as a tool to investigate texture and chemistry of lunar regolith particles

We report results of telescope polarimetric imaging of the Moon with a CCD LineScan Camera at large phase angles, near 88°. This allows measurements of the polarization degree with an absolute accuracy better than 0.3% and detection of features with polarization contrast as small as 0.1%. The measurements are carried out in two spectral bands centered near 0.65 and 0.42 μm. We suggest characterizing the lunar regolith with the parameter ( P max ) a A , where P max , A , and a are the degree of maximum polarization, albedo, and the parameter describing the linear regression of the correlation P max – A . The parameter bears significant information on the particle characteristic size and packing density of the lunar regolith. We also suggest characterizing the lunar regolith with color-ratio images obtained with a polarization filter at large phase angles. We here consider the color-ratios C | | ( 0.65 / 0.42 μm ) and C ⊥ ( 0.65 / 0.42 μm ) . Using light scattering model calculations we show that the color-ratio images obtained with a polarization filter at large phase angles suggest a new tool to study the lunar surface. In particular, it turns out that the color-ratios C | | ( 0.65 / 0.42 μm ) and C ⊥ ( 0.65 / 0.42 μm ) are sensitive to somewhat different thicknesses of the surfaces of regolith particles. We consider the applicability of the Hubble Space Telescope, the Very Large Telescope (ESO), and a spacecraft on a lunar polar orbit for polarimetric observations of the lunar surface.

Tracked pellets — a way to improve the efficiency of charmonium studies

We investigate the possibility of tracking individual hydrogen micro-spheres from an internal pellet target. Such a method aims to provide the primary vertex of a reaction to within about 100 μ m , without utilizing any detector response. Apart from background considerations the knowledge of the reaction vertex may be essential for the reconstruction of many physics channels. This is in particular true for the study of the Ψ ( 3770 ) decay into D-mesons planned at the PANDA detector at the future FAIR facility. Here, the reconstruction of displaced vertices is especially difficult since neutral particles are involved. Studies with a pellet target at The Svedberg Laboratory, Uppsala, show the technical feasibility of a tracking system utilizing fast CCD line-scan cameras. Simulations for the reaction p ¯ p → Ψ ( 3770 ) → D + D - prove the large impact such a system would have on the data taking and reconstruction at PANDA.

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

The current techniques in the textile industry cannot efficiently remove polypropylenes, packing materials and black-pigmented fibres from wool. Their presence reduces the value of the textiles considerably because they cannot be dyed and used as wool. To solve the problem, a machine vision system, consisting of a colour line-scan CCD camera, frame grabber, host computer and air-jets, is investigated and presented in this paper, which can be used to identify and remove the contaminants in scoured wool. The key techniques, the image acquisition system, algorithms and mechatronic sorting system are described in detail. Finally the performance of the machine vision system is analysed and further research work is discussed.

Analogy between soap film and gas dynamics. II. Experiments on one-dimensional motion of shock waves in soap films

This paper presents an experimental investigation of one-dimensional moving shock waves in vertical soap films. The shock waves were generated by bursting the films with a perforating spark. Images of propagating shock waves and small disturbances were recorded using a fast line scan CCD camera. An aureole and a shock wave preceding the rim of the expanding hole were clearly observed. These images are similar to the x-t diagrams in gas dynamics and give the velocities of shock and sound waves. The moving shock waves cause jumps in thickness. The variations of the induced Mach number, M2 and the ratio of film thickness across the shock wave, δ 2/δ 1, are plotted versus the shock Mach number, M s. Both results suggest that soap films are analogous to compressible gases with a specific heat ratio of γ≅1.0.

Automation in sludge dewatering by novel on-line characterisation of flocculation

A novel on-line method and system for characterisation of sludge flocs in view of improving sludge dewatering has been developed. The characterisation of sludge flocs was carried out after the conditioning or the flocculation process. The system uses a conventional CCD-line scan camera providing an on-line monitoring of the relative floc size distribution by image processing procedures. The image processing procedure has been re-evaluated and adapted to the practical dewatering results, obtained from a chamber filter press of 250 x 250 mm size. A good correlation between the calculated sensor signal and the sludge dewaterability of digested sludge in terms of the up-concentration factor was found. Although different sludge compositions and flocculation systems have been tested within the whole experimental period of six months, a good reproducibility of this correlation was also found. A well balanced floc size distribution is necessary showing that not too many but still some fine flocs and also not too large but compact flocs lead to improved dewaterability. This has been illustrated by an extended range of floc size measurements ranging between 50 microm and 29 mm. The conditioning monitoring system would be suitable for the control of production of good size-balanced flocs to compensate fluctuations in sludge characteristics of the sludge to be conditioned.

Optimum analog-to-digital converters for a laser speckle correlation-based displacement and strain sensor with optical preprocessing

We present a novel noncontact measurement procedure to determine displacements of and engineering strains within specimens at very high measurement rates. Appropriate hybrid optical and digital sig- nal processing can be employed to realize two dimensional (2D) corre- lation processing. Due to appropriate optical signal processing, the data acquisition can be performed in one dimension through CCD line-scan cameras. As a result, only 1D digital signal processing algorithms have to be executed and hence the processing power requirements decrease greatly. In this way, the measuring rates are substantially increased. It is shown that optimal Lloyd-Max quantizers implemented in the analog-to- digital converter (ADC) used in combination with a particular digital cor- relation algorithm are able to yield a further increase of the measurement rates compared with standard digital correlation techniques. © 2001 Soci-

Defect inspection in transparent materials

Describes a machine vision system which can be used for the detection of defects in glass ribbon and plastic sheets. Uses a system of multiple line scan CCD cameras and scanned LED light sources to detect and measure defects both along and across the web.

On-line surface inspection of pickled steel sheet

With the objective of controlling rejections of hot rolled flat steel, HYLSA has installed an automatic surface inspection system at the exit of the pickling line. The system is based on line-scan CCD cameras, looking at both the upper and lower faces of the steel sheet. The detection of surface anomalies and their classification into specified defect classes are carried out by the electronics of the cameras, in real time. After a given coil has been inspected, the corresponding file containing the attributes of all defects detected is analyzed off-line to test additional classification criteria and to assign a quality category. The quality assignment is based on the intensity (spatial distribution) and gravity (length and/or width) of each of the defect classes detected, as well as, on the downstream processing and the customer of each particular steel coil. The results that are being obtained look promising in regard to product grading, routing and process control.

AERODYNAMIC PARTICLE SIZE MEASUREMENT BY ELECTRODYNAMIC OSCILLATION TECHNIQUES

Four dynamical techniques for measuring the aerodynamic size of spherical and non-spherical aerosol particles are evaluated and compared. For spheres the classical method of angular light scattering is also included. One method based on particle dynamics, the so-called springpoint method, has been extensively used, but the other particle oscillation methods have been used rarely or not at all. All the dynamical methods involve imbalancing a particle in an electrodynamic balance (EDB) by changing the dc potential to produce particle oscillation. A linescan CCD camera and associated electronics were used to measure the amplitude of the oscillations, the offset of oscillation centers, and the phase lag relative to the ac drive. These measurements are compared with theoretical solutions of the equation of particle motion to establish the aerodynamic size of the particle. The stability characteristics of the particle are analyzed by solving the particle equation of motion using the method of continued fractions. The various techniques are compared for spheres, spheroids and crystalline or amorphous particles of irregular shape. All five methods are shown to be in good agreement for spheres (within 3.9%). For non-spherical germanium dioxide particles the three oscillation methods agree with the springpoint method within 3.4%.

Application of ‘vision in the loop’ for inspection of lace fabric

Within the textile manufacturing environment, inspection is an important process which is employed in order to ensure high quality of the final products. Traditionally, this process is achieved manually which is very tedious, time consuming and labour intensive. A central problem in automatic visual inspection and computer vision is to determine the extent to which one shape differs from another. This is the key element of any inspection algorithm. Pattern recognition operations such as correlation, template matching and model based vision methods can all be viewed as techniques for determining the difference between shapes. In this paper the problem of visual inspection of deformable materials in general, and lace in particular, is considered. This is particularly difficult to achieve due to inherent and inevitable variations between a ‘model’ and the material being inspected. A mechatronic approach based on correlation along with morphological filters and ‘active vision in the loop’ using a line-scan CCD camera is presented. The utilized algorithm and its advantages and disadvantages are also discussed.

Inspecting the Back of Woven Carpets for Pattern Distortion

Any distortion seen in the pattern repeat of woven Axminster carpets causes considerable problems for both manufacturer and end-user. Instrumental measurement of the straightness of the weft threads on the back of a woven carpet can be used to provide a direct indication of the nature and severity of any pattern distortion, on-line and in real time. This paper discusses two alternative methods for performing the measurement. The first, using a CCD line-scan camera and conventional digital-image-processing techniques, has been investigated fairly thoroughly and shown to be effective but is rather expensive. Evaluation of the other remains to he completed; it uses a well-spaced array of point sensors and promises to be equally effective hut much less expensive. The paper also considers how the results of the instrumental measurements should be presented to the end-user.

The automated inspection of lace using machine vision

Lace is particularly difficult to inspect using machine vision since it comprises a fine and complex pattern of threads which must be verified, on line and in real time. This paper describes instrumentation for inspecting lace actually on the knitting machine. A CCD linescan camera, synchronised to machine motions, grabs an image of the lace. Differences between this lace image and a perfect prototype image are detected by comparison methods, thresholding techniques, and finally, a neural network (to distinguish real defects from false alarms). Though produced originally in a laboratory on SUN Sparc work-stations, the processing has subsequently been implemented on a 50 MHz 486 PC look-alike. Successful operation has been demonstrated in a factory, but over a restricted width. Full width coverage awaits provision of faster processing.

Spatially Resolved Study of the Fréedericksz Transition in Thin Nematic Cells

Abstract We present a novel method for spatially resolved measurements of the amplitudes of thermal fluctuations and surface pretilt angles in thin cells filled with nematic liquid crystals (LC). We use the idea that the dynamical response of the director field during a switching process critically depends on the initial conditions of the director field which vary over the cell plane. The applied switching voltage acts as an amplifier which augments initial director deformations by orders of magnitude. By means of a CCD line scan camera combined with a polarizing microscope, we study the time dependence of the optical transmission curves. Their slopes give information on viscoelastic parameters whereas their delay can be related directly to the initial director angle.

Reverse engineering of planar parts using machine vision

Design modification of existing products or prototypes requires the extraction of the coordinate data from the part surface, especially in cases where a CAD model of the part does not exist. In these reverse engineering situations, an automated CAD model generation method could be used to accomplish the objective. This paper presents a machine vision system capable of creating a contour CAD drawing for various flat sheet metal components typical in the aerospace industry. The process of image acquisition is accomplished by the integration of a CNC machine and a CCD line scan camera. The part is mounted on the bed of the CNC machine, and the camera is fixed on the spindle of the CNC machine. The camera is set up to acquire an image of the part with the field of view of the camera covering only a small portion of the part surface. The complete surface of the part is scanned by combining several images. These images are then processed and the boundary contours are extracted. The extracted boundaries are then analyzed and combined to create a line drawing of the whole surface of the scanned part. The created data is then interfaced with a CAD/CAM system and the NC code is generated from the drawing to machine the profile and drill holes in the part. The developed algorithms and the system setup are described and sample results are presented.

Automated visual inspection of rolled metal surfaces

A prototype for an automated visual on-line metal strip inspection system is described. The system is capable of both detecting and classifying surface defects in copper alloy strips, and it has been installed for evaluation in a production line in a rolling mill. The image acquisition part of the system is based on a CCD line scan camera and condensing bright field illuminators. The inspection algorithms are based on morphological preprocessing and combined statistical and structural defect recognition. The image processing hardware consists of commercial modules. An analysis of this implementation is presented. A similar inspection principle has also been successfully applied to steel strip inspection.

Low-level-vision in an autonomous mobile robot

This paper points out the sensor system of a small mobile robot (a so called Micro Mouse), which consists of three CCD line scan cameras. The problem of the maze robot is to find the goal in the center of a maze. In contrast to the existing developments a maze robot with three CCD cameras was designed, which calculates the distance to the walls by the detection of the intensity change from the black floor to the white walls. The whole signal processing is done onboard in the small robot. This development was the base for the later construction of small laser rangefinders and at least a laser scanner, which is used in our autonomous mobile robot (MOBOT-III).