Recently, convolution neural network (CNN) has been widely used in single image super-resolution (SR). However, the traditional network structure has the problems of fewer convolution layers and slow convergence speed. In this paper, an image super-resolution method based on deep residual network is proposed. Through the deepening of the network structure, more receptive fields are obtained. Thus, more pixel information is utilized to improve the reconstruction accuracy of the model. The feature extraction process is carried out directly in low resolution space, and the images are sampled by shuffling the pixels at the end of the network. The learning method combining global residual and local residual is used to improve the convergence speed of the network while recovering the high-frequency details of the images. In order to make full use of image feature information, feature maps extracted from different residual blocks are fused. In addition, parametric rectified linear unit (PReLU) is used as the activation function, and the Adam optimization method is used to further improve the reconstruction effect. The experimental results of benchmark datasets show that the proposed method is superior to other methods in subjective visual effects and objective evaluation indicators.
In this paper, the technique of laser ultrasonic rapid detection of rail surface defects is studied, the interlaced laser ultrasonic defect detection imaging scheme is designed: the laser ultrasonic signal is excited and detected on both sides of the rail, the image is fusion-registered by algorithms such as filtering and image registration to obtain a complete rail surface inspection image to display defect characteristics, which solves the problem that laser ultrasonic is not sensitive to defects. According to the theory of thermal bomb and the proposed detection scheme, a finite element model is established to simulate the propagation process of laser ultrasonic signals in the rail, and the detection signal with surface defects information is obtained. In order to verify the effectiveness of the proposed method, a series of experiments were carried out to obtain the rail detection image with surface defects, and the influence of the laser spot size on the detection image results was analyzed. The experimental results show that the proposed laser ultrasonic imaging detection method can quickly obtain the detected image and effectively display the defect characteristics. The laser spot size has a significant influence on the detection result. When the laser spot is small, the effect of the detection image can be improved. The proposed method provides a reference for further establishing the actual rail inspection system.
The detection of rail surface defects is of great significance for railway safety. To detect the rail surface defect, the laserinduced ultrasonic rail propagation model is established by the finite element method. The intrinsic relationship between the defect depth, of the defect on rail surface and the acoustic surface wave is investigated by discussing the variation of the reflected wave and the transmitted wave both in the time and frequency domain, respectively. Quantitative evaluation of defect depth is given based on the energy of the reflected and transmitted wave, which providing a promising theoretical way for the estimation of the rail surface defect feature.
Due to the large number of points in the point cloud, the complexity of registration is quite high. To solve this problem, a registration method based on backpropagation (BP) neural network and random sphere cover set (RSCS) is proposed in this study. For the two point clouds to be registered, each is simplified based on the BP neural network. In order to avoid losing a large number of key points in the simplification process, a fixed RSCS algorithm is used for each point cloud to replace the key points with the super-point (SP) sets, and then the SP sets are combined with the simplified data. The iterative closest point (ICP) algorithm is used for fine registration. The point cloud is simplified by BP neural network and fixed RSCS, which reduce the number of points for the subsequent fine registration. Therefore, the time and space complexity can be effectively reduced. Experimental results show that the proposed method effectively improves the computational efficiency while maintaining almost the same precision details, which is of great significance for the registration of point clouds with a large number of points.
Conventional ultrasonic testing uses echo amplitude to characterize defect characteristics. Ultrasonic time-of-flight diffraction (TOFD) method detects defects by receiving the diffraction wave signal and uses the arrival time of the echo to characterize the defects. It is a highly accurate non-destructive testing method. Laser ultrasonic is a new type of noncontact ultrasonic excitation technology, which can obtain a wide frequency band signal without coupling and can simultaneously excite ultrasonic waves of various modes such as surface wave, transverse wave and longitudinal wave on the surface of the material. Among them, the surface wave and the longitudinal wave are not suitable for the TOFD detection because of their characteristics, so the transverse wave is used for defect detection in this paper. In this study, the finite element software ABAQUS was used to simulate the process of laser ultrasonic defect detection. According to the TOFD signal obtained by simulation, the size of the defect was calculated and compared with the actual size, and the detection error was obtained. At the same time, the effects of different defect length, width and depth on the echo signals are analyzed. The results show that the laser ultrasonic-TOFD method has good detection ability for defects with moderate length and small width, which indicates that it is feasible to apply TOFD method to laser ultrasonic flaw detection.
In order to avoid falling into the local optimal solution in the process of linear optimization, a calibration method for calculating the radial and tangential distortion coefficients is proposed. Firstly, the initial internal and external parameters of the camera are obtained by using the coordinates of the characteristic points in the central region of the calibration image. Camera imaging procedure is dividedinto two separate steps in which ideal image point is moved to actual image point through radial distortion and tangential distortion in sequence. With the distortion model and crossratio invariance, the distortion coefficient of the camera can be solved.. It is assumed that the point after distortion correction conforms to perspective projection principle, and the exact value is approximated step with the distortion model and iteration. Compared with the traditional nonlinear optimization method, the average back-projection error of the corrected image coordinates is significantly reduced and has better robustness and accuracy.
It is beneficial for maintenance department to make maintenance strategy and reduce maintenance cost to forecast the hidden danger index value. In order to grasp the size information of High-speed railway wheel-set size in time and ensure the stable operation of high-speed railway, the size data of wheel-set are obtained by optical intercept image detection, and the LMBP neural network prediction model based on differential evolution is designed and implemented. The differential evolution algorithm (DE) is used to optimize the initial connection weights and thresholds between the layers of the neural network, and solve the problem that the back propagation (BP) network is easy to fall into the local extreme value due to the random initial connection weight and threshold. The Levenberg-Marquardt (LM) numerical algorithm is used to optimize the weights and thresholds in the BP network training process to solve the problem of long BP training time. According to the wheel diameter data of the CRH380 model, the effectiveness and accuracy of the method are verified by comparing the prediction results of different algorithms. Compared with the LMBP network and the standard BP network prediction model, the experimental results show that the DE-LMBP neural network model can obtain better correlation coefficients (0.9974), mean square error (0.0103), mean absolute error (0.0772) and average absolute percentage error (0.0084), which proves that the model is effective in predicting the size of the moving wheel and significantly improves the prediction accuracy.
Proc. SPIE. 10843, 9th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Sensing and Imaging
Edge detection is a crucial task in image processing. Owing to the similarity in property between edges and noise, which demonstrates abrupt changes in image grayscale values, traditional edge detection methods are insufficient in detecting weak edges. Therefore, a local multi-threshold fuzzy inference method (LMFI) is introduced. Considering the binarization processing prior to conducting a fuzzy inference, to retain more edge information, a local threshold processing method and a triple threshold processing method are proposed. To reduce noise interference, an improved sigma filter and an improved fuzzy inference strategy are presented. The experimental results show that the effect of weak edge detection is improved by LMFI, when compared to conventional methods such as the original fuzzy inference algorithm and Canny edge detection algorithm.
Proc. SPIE. 10964, Tenth International Conference on Information Optics and Photonics
KEYWORDS: Fringe analysis, Logic, Clocks, Video, Field programmable gate arrays, LCDs, 3D metrology, Projection systems, Signal generators, Digital Light Processing
This paper presents a method about fringe pattern generation of three-dimensional (3D) shape measurement based on field programmable gate array (FPGA). The system hardware is described by logic circuits in Verilog hardware description language (HDL). The system includes signal generation module, signal control module, fringe pattern selection module, video graphics array (VGA) and liquid crystal display (LCD) module. FPGA is used for generating fringe patterns with randomly adjustable frequency, phase and type in the system. Afterward, the fringe pattern is projected onto the object under test by the digital light processing (DLP) projector. The fringe pattern generated by the system is stable and accurate, not affected by the environment and space of measurement. At the same time, it can also make up for the shortcomings of the traditional measurement methods which rely on the computer generating fringe pattern. This method not only cuts down size and cost of the system, but also improves the measurement quality.
X-ray testing is based on the attenuation of X-rays when passing through matter. Image detectors acquire the X-ray information which is defined by the local penetrated wall thickness of the tested sample. By X-ray absorption in the detector and following read-out and digitization steps a digital image is generated. As detectors a radiographic film and film digitization, a storage phosphor imaging plate and a special Laser scanner (Computer Radiography - CR) or a digital detector array (DDA) can be used. The digital image in the computer can then be further analyzed using many types of image processing. In the presented work the automated evaluation of wall thickness profiles are investigated using a test steel pipe with 9 different wall thicknesses and various X-ray voltages and different filter materials at the tube port and intermediate between object and detector. In this way the influence of different radiation qualities on the accuracy of the automated wall thickness evaluation depending on the penetrated wall thickness of the steel pipe was investigated.
Atmospheric anisoplanatic effect is an important problem to be solved in telescope observation of space target imaging. Numerical simulation of atmospheric anisoplanatic imaging is the basis for studying the restoration of anisoplanatic images. Based on the propagation theory of light waves on the inhomogeneous turbulent path and multilayer phase screens distribution model, this paper establishes a theoretical model of atmospheric imaging for space targets under anisoplanatic conditions. The near-surface atmosphere can be divided into several stratifications of atmosphere at different altitudes. Find out the best phase screen distribution location for each atmospheric stratification, and use the multilayer phase screens at different altitudes to represent the atmospheric anisoplanatic effect. The phase change of the light wave emitted by each point on the space object through the atmosphere is represented by a phase screen, and the final phase size is the superposition of the phase of the light wave passing through the phase screens of each layer. A series of spatial target images are simulated by different layers of phase screens for anisoplanatic imaging, and combined with theoretical analysis to find the best phase screen position and the number of layers. The experimental results show that the three-layer phase screen can accurately simulate the atmospheric anisoplanatic imaging while maintaining the computational efficiency, and effectively reflect the changes of the point spread function (PSF) when the spatial position changes. The imaging results have no ringing and edge effects, and can accurately represent the influence of atmospheric anisoplanatic effect on atmospheric imaging.
Matching accuracy plays a vital role in image matching and image recognition. In order to improve the accuracy and robust of image matching, a bidirectional matching algorithm is proposed to delete false matching relationships, so the matching accuracy can be improved. Based on the unique constraint of unilateral matching, the positive matching results from template image to the image to be matched can be obtained. Then the negative matching results from the original image to be matched to the original template image can also be obtained. Now the final bidirectional matching results can be achieved by the intersection of the positive and negative results. Precision ratio is taken as the evaluation indicator. Through various image transformation scenes, experimental results show that the proposed algorithm has a higher precision ratio on the contrast of unilateral matching algorithm. So the proposed bidirectional matching algorithm can improve the precision ratio and robust of unilateral matching algorithm and improve the matching accuracy of image matching.
Proc. SPIE. 9684, 8th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test, Measurement Technology, and Equipment
KEYWORDS: Image fusion, Fringe analysis, Statistical analysis, 3D image reconstruction, Error analysis, Computer simulations, 3D metrology, Reconstruction algorithms, 3D image processing, Phase shifts
PMP (Phase measuring Profilometry) is an excellent 3D online measurement method for its high precision. However, the
measuring range is limited. While the rail is so long that far exceeds the measuring limit, the image stitching should be
used to extent it. In this paper, based on the improved Stoilov algorithm, the rail shape is three-dimensionally
reconstructed and the abrasion is detected combines image stitching. Two types of schemes are researched: (1)image
stitching is firstly used on the deformed fringe patterns and then a larger range rail is constructed with Stoilov algorithm;
(2)the three-dimensional construction of two fringe pattern is firstly performed, and then the constructed images are
stitched into longer rail. In this paper, the improved Stoilov algorithm based on statistical approach and stitching
algorithm are analyzed. 3D Peaks function is simulated to verify the two methods, and then three-dimensional rail shape
is recovered based on these two methods and the rail abrasion is measured with the relative precision of higher than
0.1%, which is much higher than traditional methods, such as linear laser scanning.
Proc. SPIE. 9684, 8th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test, Measurement Technology, and Equipment
With the rapid development of high-speed and heavy-load in modern rail transit, the abrasion and surface defect of rail
are getting serious, and the demand of measuring the rail shape and surface defect has been rising. Phase Measuring
Profilometry (PMP), due to the good characters of non-contact, high precision, easy to control automatically etc., is often
used for precise 3D shape reconstruction. In this paper, PMP technology and Stoilov phase shift algorithm are adopted,
three deformed fringe patterns of rail are collected with fixed phase shift between them, and branch cut phase
unwrapping algorithm is used, based on which the three-dimensional surface shape of the rail is reconstructed and the
artificial surface flaws are restored and measured. This method provides a good reference for the precise online detection
of the rail abrasion and surface defect.
Proc. SPIE. 9282, 7th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment
KEYWORDS: Optical filters, Modulation, Image processing, Digital filtering, Wavelets, Fourier transforms, Computer simulations, 3D metrology, Spatial resolution, Continuous wavelet transforms
In three dimensional detection of the rail shape by Fourier Transform profilometry (FTP), filtering is one of the key links before Fourier transform. The choice of filtering window decides the spectrum overlapping degree of deformed fringes, so as to decide the measurement precision of the rail shape. In this paper, based on wavelet ridge theory the size of the filter window is self-adaptive according to the frequency alternation of deformed fringes. And thus the optimum matching window size is decided, the frequency overlapping is furthest reduced and the measurement precision is improved. Simulation and experiments manifest that self-adaptive filtering can greatly enhance the precision in three dimensional detection, which offers a new thinking and method in rail shape recovery and defect detection.
The defects of wheel sets seriously affect the train operation safety which of problem to be solved. A method of
detecting the wheel sets defect is investigated through the three-dimension profile. The three-dimension profile
reconstruction of wheel sets is realized by combining two-dimension CCD imaging with line laser and encoder. The
theoretical derivation of detecting principle is established, and the factors of influencing the measurement accuracy are
analyzed. An algorithm is set to processing measurement data, the main parameters of wheel set are obtained, including
flange thickness, flange height, vertical wear (QR) and tread wear. The results of simulations and field experiments show
that the proposed method can detect the faults on the wheel correctly, and satisfy the requirements of high efficient and
accurate.
A novel approach is proposed for obtaining high resolution image of removing the optical aberrations by disturbing the
optical wave-front phase and digital image processing. An optical random phase mask of the phase spectrum
fluctuation corresponds to Kolmogorv distribution is placed between the exit pupil and image plane of optical system to
make the optical aberration image blurred termed the intermediate image. The intermediate image acquired by digital
detector is restored through the blind deconvolution algorithm base on maximum-likelihood estimation technique. The
effect of optical aberrations on restoration image and superresolution performance of image was explored. As a
demonstration to verify the utility of this method, the primary aberrations corresponding to the optical system are
applied, and the image of removing the aberrations by a computer simulation and experiment is shown. The results
suggest that the present method is well suited for improving the imaging quality of the optical system, and partly
removing the diffraction effect of optical system on restoration image.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print format on
SPIE.org.