Detecting parabolas in ultrasound B-scan images with genetic-based inverse voting Hough transform (original) (raw)
Related papers
Crack Defect Detection and Localization Using Genetic-Based Inverse Voting Hough Transform
2002
In this paper we propose a Genetic-Based Inverse Voting Hough Transform (GBIVHT) method to detect buried crack defects in engineering structures. The method is applied to B-scan images obtained according to the ultrasonic Time Of Flight Diffraction technique. In these image representations of the ultrasound data, crack defects are characterized by multiple arcs of diffraction that can be approximated by a parabolic model. Thus, the crack defect detection problem in non-destructive inspection of engineering structures is transformed into a parabola detection and localization on B-scan images. In the proposed GBIVHT method, the local peak detection problem of conventional HT is converted into a parameter optimization problem that operates directly on the B-scan images. The optimization task is done using the wellknown Genetic Algorithms. Our main goals are an accurate detection of the parabolas while circumventing the computational complexity and huge storage problem tied to conventional HT.
Detection of cracks in materials using the randomized Hough transform on ultrasonic images
Proceedings of the 6th WSEAS …, 2006
Time Of Flight Diffraction (TOFD) is a non destructive testing (ndt) technique which use the diffraction of the ultrasonic waves on the tips of discontinuities. In this technique, the data are displayed in the form of images and processing algorithms can be applied , allowing thus automatic detection and characterisation of defects. The work we present here is an image processing algorithm which permits to detect and locate automatically several cracks present in a structure by analysing a TOFD image. This algorithm consists of two phases : In the first one , a pre-processing reduces the image to a sparse matrix containing only significant pixels , and separates the different groups by a graph-partioning . The second phase (the decision one ) is an application of the Randomized Hough Transform on each population of pixels in order to detect and locate parabola characterising each crack defect .
A Real-coded Genetic Algorithm for Identification of Defects with Ultrasound Time-of-Flight Data
Lecture Notes in Networks and Systems , 2022
Tomographic reconstruction enables to look through engineering objects for material defects without hindering the objects functionality. Tomographic reconstruction procedures available in literature are developed in the lines of binary genetic algorithm principles. The binary genetic algorithm procedure reported earlier required the user to input the actual values of characteristic properties of materials present in an objects cross-section. The previous algorithm was designed to estimate the size and location of the defects, given the characteristic properties of object and defect material. In this work, a tomographic reconstruction procedure is modeled and implemented using the principles of real genetic algorithms. The algorithm is shown to achieve the task of reconstructing the cross-section of a test specimen starting with initial guesses that are within a given range of characteristic property to be estimated. The present algorithm does not necessitate the user to input exact characteristic property of material defects assumed to be present in the material cross-section being examined. The efficacy of proposed reconstruction algorithm is demonstrated with several numerical simulation studies.
Detection of Inclusions using the Hough Transform on Ultrasonic Image
In non destructive testing of materials , the recourse to the imagery allows, more than a convivial representation of the results , to carry out automatically the operations of detection , localisation and sizing of defects eventually present in a structure. The rapidity of the calculators and their graphic performances permit to use today classical tools of image processing which need voluminous quantities of calculations . Our work consists of applying the Hough transform to detect inclusions in a material by analysing an ultrasonic image type c-scan . In a first step , this image is binarized using the -6dB method which reduce the defects to their real sizes . The Hough Transform is then applied on the edges in order to detect circular forms characterising the inclusions.
Application of genetic algorithms to ultrasonic tomography
The Journal of the Acoustical Society of America, 1998
A new method for ultrasonic tomography based on genetic algorithms is proposed for the prediction of the geometry of an inclusion of known physical properties in a given specimen. New inversion operators are introduced in order to take full advantage of the physical properties of the system investigated. The efficiency of the proposed method is tested through a comparison with other existing techniques for the solution of the inverse problem. The accuracy of the procedure is verified by using a variety of sets of synthetic data. Good and fast convergence is obtained even in the case of complex geometries if parallel processing is adopted.
Applicability of genetic algorithms to reconstruction of projected data from ultrasonic tomography
Proceedings of the 10th annual conference on Genetic and evolutionary computation - GECCO '08, 2008
In this paper simulation studies of the ultrasound computerized tomography (CT) technique employing time of flight data is presented. An enhanced genetic algorithm based reconstruction technique is proposed that is capable of detecting multiple types of inclusions in the test specimen to be reconstructed. It is assumed that the physical properties of the inclusions are known a priori. The preliminary results of our algorithm for a simple configuration are found to be better than those reported with MART1. In addition to being able to identify inclusions of different materials, both the shape and location of the inclusions could be reconstructed using the proposed algorithm. The results are found to be consistent and satisfactory for a wide range of grid sizes and geometries of inclusion(s). Based on the regression analysis an empirical relation between the number of unknowns and the reconstruction time is found which enables one to predict the reconstruction time for higher resolutions.
2020
Engineering materials and structures have crack-like defects leading to premature failures. Usage of fracture mechanics to assess the structural integrity requires knowledge on the type, location, shape, size, and orientation of the flaws. Tomographic reconstruction is one of the commonly used nondestructive testing methods for flaw characterization. The cross sectional image of the object being tested is obtained through the application of various reconstruction methods that are categorized as either analytical methods or iterative methods. In this work an iterative algorithm that works on the principles of genetic algorithms is developed and used for the reconstruction. The results of simulation studies on the tomographic reconstructions using genetic algorithms for the identification of defects in isotropic materials are discussed in the paper. The solution methodology based on use of genetic algorithms is applied to reconstruct the cross sections of test specimens with different...
Evolutionary improved object detector for ultrasound images
2013 36th International Conference on Telecommunications and Signal Processing (TSP), 2013
Object detection in ultrasound images is difficult problem mainly because of relatively low signal-to-noise ratio. This paper deals with object detection in the noisy ultrasound images using modified version of Viola-Jones object detector. The method describes detection of carotid artery longitudinal section in ultrasound B-mode images. The detector is primarily trained by AdaBoost algorithm and uses a cascade of Haar-like features as a classifier. The main contribution of this paper is a method for detection of carotid artery longitudinal section. This method creates cascade of classifiers automatically using genetic algorithms. We also created post-processing method that marks position of artery in the image. The proposed method was released as open-source software. Resulting detector achieved accuracy 96.29 %. When compared to SVM classification enlarged with RANSAC (RANdom SAmple Consensus) method that was used for detection of carotid artery longitudinal section, works our method real-time.
New Algorithm Based on S-Transform to Increase Defect Resolution Within Ultrasonic Images
Recent Advances in Engineering Mathematics and Physics, 2020
Recent years have seen a notable advance in the quality of produced industrial ultrasonic data. This is due to two main factors. On the one hand, advances on hardware level permitted to design and trigger ultrasound sensing arrays and matrices that led to new acquisition strategies such as phased array method and full matrix capture technique. On the other hand, the development of algorithms and software components to reconstruct and process the measured data allowed a major improvement of the signal and image quality. Within this aspect, modern signal processing algorithms improved the defect resolution and thus their detection in ultrasound data. Mostly, methods based on time-frequency analysis are used. The measure of the improvement resulting from the signal processing methodology can be confirmed, for instance, by evaluating A-scans containing defects near the front and the back wall of inspected specimens. In this work, we describe a novel algorithm for processing one-, two-or three-dimensional ultrasonic data, in order to increase their defect resolution. The algorithm is demonstrated using simulation phantom as well as on a real specimen both including defects at different depths. The proposed enhancement method is based on the Stockwell transform and normalized
Theoretical and Applied Fracture Mechanics, 2008
A hybrid-GA method, based on signal parameterization, has been reported here for the improved detection and sizing of surface cracks of small sizes/depths in thin sections. The method relies on parameterizing the composite reference from the defect into its individual components i.e., the crack tip diffracted echo and the corner trap echo and subsequently use the relative arrival time technique (RATT). The phased array ultrasonic technique was employed in the investigation. Both experimental and simulated signals were used in the study. It is shown through both simulations and experiments that the hybrid-GA is successful in parameterizing both non-overlapping and overlapping echoes encountered in thin sections. It is additionally shown that the hybrid-GA improves the signal to noise ratio and correct for under-sampling of data.