This paper describes the development and system integration of a non-cooled industrial thermal imaging system (ITIS) based on an uncooled linear array 1 X 128 pyroelectric detectors and internal scanner and the concept of multisensor industrial thermal imaging system for inspection/monitoring thermal processes or hazardous environments. In previous time, as first step, was developed a thermal imaging system in which was a two dimensional capability is reached with mounting of the camera on the precise programmable positional unit that exhibits positioning capability in 2 D.o.F. (degree of freedom). As a second stage of a developing an industrial thermal imaging system was mechanical design and construction of an internal scanner for linear array. Internal scanner has ability to create two-dimensional frame with selectable numbers of rows (from 1 to 128). Frame rate is depending on a numbers of rows in a frame (from 128 Hz to 1 Hz) and software averaging. The measured NETD of a thermal imaging system, without frame averaging, is better than 0.5 degrees Celsius (for lens with F equals 0.7 at 300 K). The digital part of a camera is equipped with 32 bit floating point digital signal processor (SHARC, Analog Devices), fieldbus controller (COM20020, SMCS, with implemented low-level protocol ARCNET ANSI ATA 878.1 with data rate 5 Mbit per second) and fieldbus interfaces (for twisted pair and coaxial cable). The positional unit was also attached with fieldbus controller. This paper also deals with a conceptual design of a multi- sensor and multi-actuator system based on a ARCNET Local Area Networks (LAN) with set of nodes like thermal or visible cameras, positional units and other necessary nodes.
Conventional thermal wave imaging methods based on optical excitation are sensitive to all thermal boundaries and structures. However, if ultrasound is used for energy deposition, areas of enhanced mechanical loss angle are heated up selectively. As defects are usually correlated with an enhanced loss angle, ultrasound lockin thermography (ULT) is a defect-selective NDE method. Thermal wave emission from defects is monitored that are activated by absorption of ultrasound. In this paper we investigate the applicability of ULT for maintenance and inspection. Examples are presented showing the detection of defects in thick composite materials, of vertical cracks in several materials, and of hidden corrosion. Limitations given by the use of ultrasound are also discussed.
Lockin thermography is currently being used for the rapid and remote identification of subsurface structures and defects such as impact damages, delaminations, and hidden corrosion. The purpose of this paper is to show that lockin thermography is also a reliable tool to inspect in a remote way the tightness of mechanical joints in safety relevant structures (e.g. aerospace equipment and vehicles). For example, the rapid identification of loose rivets is a major concern for airlines and manufacturers in order to monitor the structural integrity of their aircraft. Our measurements aimed at the early detection of loose rivets. We analyzed the phase image signature obtained on two metal plates pressed together by screws fastened at various torque levels. A clear relationship was established between phase angle and torque level at which the screws had been fastened. Based on these results our investigations were extended to riveted samples where two aluminum plates were pressed together by an array of ten blind rivets. Also in this case the level of tightness of rivet joints can be detected. In addition to these feasibility studies on model samples, we performed investigations on airplane components which confirmed the applicability of lockin thermography for remote maintenance inspection within a few minutes.
The presence of moisture in building materials causes damage second only to structural one. NDT are successfully applied to map moisture distribution, to localize the source of water and to determine microclimatic conditions. IR Thermography has the advantage of non-destructive testing while it allows to investigate large surfaces. The measures can be repeated in time to monitor the phenomenon of raising water. Nevertheless the investigation of moisture in walls is one of the less reliable application of Thermography IR applied to cultural heritage preservation. The temperature of the damp areas can be colder than dry ones, because of surface evaporation, or can be warmer, because of the higher thermal inertia of water content versus building materials. The apparent discrepancies between the two results are due to the different microclimatic conditions of the scanning. Aim of the paper is to describe optimal procedures to obtain reliable maps of moisture in building materials, at different environmental and microclimatic conditions. Another goal is the description of the related energetic phenomena, which cause temperature discontinuities, and that are detected by thermography. Active and passive procedures are presented and compared. Case studies show some examples of procedures application.
The cause of much damages in buildings, particularly in old buildings, is a direct or indirect consequence of moisture. For diagnosis of moisture in building materials it is necessary to perform remote measurements non-destructively with high selectivity. The method should be able to locate large areas with excessive moisture. For this presentation of the problem we developed the amplitude sensitive modulation thermography. This method utilizes the absorption behaviour of water in the wavelength of 1 .94 zm and a reference wavelength band at 1 .0 rim. To detect only this two wavelength bands without the influence ofleaking radiation, a modulated illumination is used. For a better comparability and an elimination of measuring effects we generate from the reference- and the measuring effect wavelength band a surface moisture image. With a preliminary calibration in laboratory the quantitative detection of surface moisture is achieved. The results of this investigation eliminate the influence of leaking radiation, geometric and surface conditions effects by reducing emission behaviour of the surface. At the moment the measurements present a trend to measure the moisture in different materials without a material characteristic dependence.
This paper presents an experimental system for machine vision guided waterjet cutting. Machine vision is used to provide accurate real-time data on the outline and position of the preform utilised in the nesting process, leading to material and cost savings. The experimental system comprises a computer-controlled waterjet cutting system, a machine vision system and a CAD system. Three 5000-pixel linescan cameras are employed in the vision system, providing a spatial resolution of 0.2 mm over the 3 m wide table. Image analysis software, height correction and calibration methods have been developed for the application, and these have been tested under real conditions.
A fast and efficient method for the detection and recognition of objects which have similar, but not identical, contours is presented. Arbitrary shapes are characterized by interpreting their boundary points as complex numbers and generating spectra from those representations using the Fast Fourier Transform. Suitable normalization of those spectral components leads to a translation, scale and rotation invariant description of each shape. A similarity measure which is based on a simple distance calculation of the spectral magnitude components is used to classify each new shape. By selecting a specific number of spectral components (lower frequencies describe coarse obj ect details, higher frequencies explain fine details) the whole recognition process may be easily tailored to specific needs of recognition accuracy, performance and allowed shape deviation. Besides the compactness of object description, our proposed algorithm for shape recognition can be very efficiently implemented and executed in real-time on standard PC hardware
In this research characteristics of standard commercial CCD and CMOS cameras are evaluated experimentally and compared. Special attention is paid to the operation of these devices in low light level condition, which is typical to many surveillance and consumer electronics applications. One emerging application utilizing inexpensive image sensors at variable illumination condition is the UMTS (Universal Mobile Telecommunications System), which will deliver wirelessly, for example, pictures, graphics and video from the year 2002. The determination of the system performance is based in this study on the imaging of a calibrated gray scale test chart at varying illumination condition. At each level of illumination the system response is characterized by a signal to random noise figure. The signal is calculated as the difference of the system response to the lightest and darkest areas of the gray scale. The random noise is measured as the standard deviation of the gray values in a difference of two successive images of the test pattern. The standard deviation is calculated from 10-bit digitized images for small group of pixels (36 X 36) corresponding to the different areas of the gray scale in the test pattern images. If the random noise is plot as a function of signal (encoded in digital numbers, DN) for small group of pixels, a Photon Transfer curve is obtained. This is one of the basic performance standards of CCD sensors. However, if camera systems with nonlinear response or AGC are evaluated, the variations of the system response at different signal levels should be included to the performance measure. In these cases the signal to noise curve is useful. The signal to random noise curves were determined for a CCD and a CMOS camera characterized by similar specifications. The comparison between two camera systems shows that considerable differences between the operation of these devices especially at low light level condition can exist. It was found that approximately a sevenfold illumination level is needed in the case of the exemplary CMOS camera compared to the CCD camera in order to achieve a reasonable imaging performance.
In particle size measurement by imaging methods, two types of illumination are encountered: coherent and non-coherent. For both the analyzed volume is limited by out of focus effects. The aim of our study is to turn away that limitation by a morphological analysis of particle images obtained with a configuration close to in-line Gabor holography. Fringe pattern appears in these images and it is no longer comparable to non-coherent shadowgraphy images which become fuzzy with out of focus positions of the object. A simulation of the image formation has been performed thanks to a model. The diameter of the spherical particle and its z position from the focus plane are parameters of the model. Particle sizing will be derived from original morphologic criteria inferred from the fringes. These criteria can be defined and validated by simulation and experimental results. This paper presents: the model and its characteristics, a comparison of experimental and simulated images, and the morphological analysis for different particle sizes and positions.
Interferometric methods are of particular interest in industrial quality assurance to detect material defects via deformation and vibration analysis. One of their main advantages is contact-free surface examination and the possibility to provide full-field information with sub-micron sensitivity. When examining intensity images provided by a phase-shifting speckle interferometer, the measuring range is limited because the phase wrapping effect occurs in combination with high speckle noise levels. In this paper, a new hybrid approach for the phase unwrapping problem is presented which significantly extends the measuring range, and, in addition, reduces the computational effort. This is achieved by a specific tile processing strategy combining a polynomial based local phase-unwrapping approach and a Fourier based strategy to form a hybrid algorithm. The hybrid approach assures time efficient tile processing using polynomial approximation in image areas of low fringe density on the one hand and carefully directed high performance slope computations in noisy or strongly deformed areas on the other hand. In these strongly deformed areas a Fourier based approach can still find the correct affine linear approximation. With the resulting set of local affine linear approximations, a global spline model of the underlying deformation profile is obtained by an iterative region-growing technique combined with a O(n) data transformation.
Matching of a reference template with an image is a computationally expensive job. Particularly in fast real-time applications, large images and search ranges led to serious implementation problems. Therefore a reduction of the template size achieved by the selection of an appropriate subtemplate which is used for point correlation (subtemplate matching) may significantly decrease computational cost. In this paper a modified algorithm of the subtemplate point selection is proposed and explored. With the additional use of image pyramids, we can reduce the computational costs even further. The algorithm starts with a coarse search grid in the top level of the image pyramid generated for the full intended resolution. The procedure continues until the lowest level of the pyramid, the original image, is reached. The computational costs of this algorithm part satisfy the requirement for on- line processing. The preparation of the subtemplate for the point correlation is carried out in off-line mode, i.e., there is no rigorous limit of computational costs. The technique that applies the point correlation to image template matching within the image pyramid concept is proposed and the results obtained are discussed. It is especially useful for fast real- time system implementation when a large number of template matchings are needed in the same image.
Ground Penetrating Radar (GPR) is increasingly used for non- destructive investigations of concrete structures in civil engineering. The success of any radar investigation depends on the duration of the radiated radar impulse and on the radiation pattern of the antenna. These antenna characteristics influence the spatial resolution of the radar image. To increase the spatial resolution, a high frequency antenna was developed. For this new antenna and two commercial antennas both mentioned antenna characteristics were determined using a concrete half sphere. The different performance of these antennas due to the different characteristics is demonstrated by comparison measurements at two concrete specimens.
In a ventilated space, the incoming air jet and the resulting air flow pattern play key roles in the removal or supply of heat, moisture, and harmful gases from or to living organisms (man, animal, plant). In this research, two low cost image processing methods (boundary peeling and optical flow) were developed to quantify the 2-D trajectory and the deflection angle of an air jet in a ventilated room. The image processing algorithms were applied to a wide range of air jets in a laboratory test room. Based on the deflection angle of the air jet trajectory from the initial path, a distinction could be made between falling, instable or horizontal air jets.
The application is developed for measuring the dimensions and the three dimensional shape of soap stone bricks. The bricks are used as elements of stone ovens. Each brick is approximately 280 millimeters wide, 70 millimeters thick and the length varies from 140 up to 630 millimeters. The shape of a stone is measured by two images captured from known camera positions by comparing the images with the projected images of the floating 3D-model. The original images and the projected images from the model are fitted together by changing the shape and the position of 3D-model. An unknown stone can be measured by finding the optimal shape and the position of the model in relation to the original image data of that stone. In other words, the synthetic stone is moving, rotating, and reshaping between two fixed cameras, looking for the best fit to the original image data.
Method for 3D reconstruction of known rigid objects from a single monocular image or a sequence of monocular images is presented. In the first part of the paper, a new computational approach to estimate pose and orientation of well-known objects in a 3D-scene from a single frame is discussed. The underlying theory is described in the context of a prototype matching problem and the existence of optimal solutions is proved. Furthermore, it is straightforward to extend the concept of prototype matching to the case of stereo or pseudo stereo applications and even more general setups. Hence, an improved 3D reconstruction with higher accuracy and increased stability can be achieved, for instance, by moving the camera along a linear sledge. The estimation of the ego-motion of a camera is covered as a special case by the introduced modeling via prototype matching. In the second part, experimental results for natural image sequences are analyzed. The derived accuracy and execution times of the described algorithms are illustrated. The last section deals with implementational details necessary to reduce the execution time under real-time restrictions.
Within the European Mars Express Mission to be launched 2003 the Beagle2 Lander will foresee the access to stereoscopic views of the surrounding Martian surface after touchdown. For scientific purposes the necessity for a high resolution three dimensional (3D) reconstruction of the landing site is evident. A lander vision subsystem capable of reconstructing the landing site and its vicinity using a stereo camera mounted on the robotic arm of the lander is used therefore. Knowledge about the geometric camera features (position and pointing with respect to each other, position and pointing with respect to the lander, intrinsic parameters and lens distortion) are determined in a calibration step on ground before takeoff. The 3D reconstruction of the landing site is performed after landing by means of stereo matching using the transmitted images. Merging several stereo reconstructions uses the respective robotic arm states during image acquisition for calibration. This paper describes the full processing chain consisting of calibration of the sensor system, stereo matching, 3D reconstruction and merging of results. Emphasis is laid on the stereo reconstruction step. A software system configuration is proposed. Tests using Mars Pathfinder images as example data show the feasibility of the approach and give accuracy estimations.
In this text a novel active infrared thermography approach is proposed for land mine detection. The simple approach consists in spraying the soil to inspect with water having a different temperature than soil temperature. Different temperature zones witness presence of buried mines. Theory related to pulsed thermography, simple simulation of the process, some results and envisaged battlefield deployment are presented.
Different coding allows for the representation of the signals in a digital form with small word length and at the same time it ensures a specified resolution. The use of differences and delta modulation (DM) for the economical representation and efficient processing of the images needs the development of the existing methods and the working out of simple and fast processing methods in real time. One of the perspective processing methods is the unclear extreme analysis (UEA). Methods of signal compression and reconstruction (in the real time mode) with the use of UEA and DM are simple, economical and fast. The dependence of degree compression and a specified accuracy of the signal reconstruction with the use of UEA, first and secondary 2-D differences, separately and together, are showed in the examples. The influence of the limits of unclearance for them is shown too. This type of compression and reconstruction can be realized with the help of economical and technological devices, especially with the homogeneous computing environments.