Surface Detection Algorithm Research Articles

Analyze Gear Failures and Identify Defects in Gear System for Vehicles Using Digital Image Processing

Gear defects are a major reason for poor quality and of embarrassment for manufacturers. 0 Inspection processes done on these industries are mostly manual and time consuming. To reduce error on identifying gear defects requires more automotive and accurate inspection process. Considering this lacking, this research implements a Gear Defect Recognizer which uses computer vision methodology with the combination of local thresholding to identify possible defects. The recognizer identifies the gear defects within economical cost and produces less error prone inspection system in real time. In order to generate data set, primarily the recognizer captures digital gear images by image acquisition device and converts the RGB images into binary images by restoration process and local threshold techniques. Later, the outputs of the processed image are the area of the faulty portion and compute the possible defective and non -defective gear as an output. Detection of bad quality plastic gears is critical for any manufacturing unit trying to make a mark in the market in terms of quality standard and cost. Here we explore the possibility of using image segmentation and algorithms like non-smooth surface detection algorithms to automate the process of defect detection. In these plastics we have picked industrial strength plastic gears used typically in applications like robotic arms where quality in paramount for the functioning of the device. In this paper review of various gear defects and the possible automated solutions using image processing techniques for defect detection is given.

Real-time surface tracking system using common-path spectral domain optical coherence tomography

An enhanced surface tracking system based on optical coherence tomography (OCT) modality has been developed and tested for use in a surgical guidance system. A surface detection algorithm based on a Savitzky-Golay filter of A-scan data and thresholding was applied to real-time depth tracking. The algorithm output controlled a motorized stage to adjust the probe position according to the sample's topological variance in real-time. As a result, the root mean square error (RMSE: 4.2 μm) of our algorithm was relatively lower than the conventional method (RMSE: 16.6 μm). Also, OCT images obtained using the algorithm showed a significantly extended imaging range and active surface tracking in real time. Consequently, the devised method demonstrated potential for use in systems for guiding surgical robots and endoscopic OCT.

A stabilized Runge–Kutta, Taylor smoothed particle hydrodynamics algorithm for large deformation problems in dynamics

SUMMARYNumerical simulation of large deformation and failure problems present a series of difficulties when solved using mesh based methods. Meshless methods present an interesting alternative that has been explored in the past years by researchers. Here we propose a Runge–Kutta Taylor SPH model based on formulating the dynamic problem as a set of first‐order PDEs. Two sets of nodes are used for time steps n and n + 1 ∕ 2, resulting on avoiding the classical tensile instability of some other SPH formulations. To improve the accuracy and stability of the algorithm, the Taylor expansion in time of the advective terms is combined with a Runge–Kutta integration of the sources. Finally, as boundaries change during the process, a free surface detection algorithm is introduced. Copyright © 2012 John Wiley & Sons, Ltd.

Combining Boundary and Region Information with Bolt Prior for Rail Surface Detection

Railway inspection is important in railway maintenance. There are several tasks in railway inspection, e.g., defect detection and bolt detection. For those inspection tasks, the detection of rail surface is a fundamental and key issue. In order to detect rail defects and missing bolts, one must know the exact location of the rail surface. To deal with this problem, we propose an efficient Rail Surface Detection (RSD) algorithm that combines boundary and region information in a uniform formulation. Moreover, we reevaluate the rail location by introducing the top down information — bolt location prior. The experimental results show that the proposed algorithm can detect the rail surface efficiently.

TH-C-201C-03: Tumor Segmentation in CT Images Using Globally Optimal Single Surface Detection

Introduction The problem of accurate and reproducible tumor definition is essential for radiation therapy in treatment planning as well as for volumetric assessment of tumor response to therapy. To improve the quality and reproducibility of the tumor definition as well as to accelerate the workflow of the oncologist and radiologist novel automated segmentation tools are needed. Nodule and lesion segmentation remain challenging and are not yet solved satisfactorily. Method and Material Our tumor segmentation algorithm is based on a single surface detection algorithm using regional properties. To use this method for tumor segmentation we created a framework which requires a small amount of user input that is then followed by an ellipsoidal transform on the data. The framework then uses the single surface detection method optimizing the intra‐class variance. The surface detection method utilizes the techniques of shape probing graph search and parametric maximum‐flow. Our algorithm was tested for the segmentation of livertumors from 15 CTimagedata sets and lungtumors from 18 CTimagedata sets.Surface positioning error and volume measures compared with expert‐traced results were computed. Results Our segmentation method demonstrated low surface positioning errors and robust performance compared with the expert‐traced results. The average signed and unsigned positioning errors for liver lesions were −0.07±0.31 and 0.77±0.16 voxels. For lungtumors the average signed and unsigned positioning errors were −0.03±0.10 and 0.78±0.13 voxels. Volume measures also showed accurate and robust correlation with expert‐traced results with 90% of acceptable percentage in terms of volume overlap. Conclusion We have implemented a novel single surface detection method that minimizes the intra‐class variance and provided a framework for tumor segmentation. Experiments on liver lesions and lungtumors show good applicability with sub‐voxel accuracy achieved for both cases.

Measurement of Left Ventricular Mass by Real-Time Three-Dimensional Echocardiography: Validation Against Magnetic Resonance and Comparison with Two-Dimensional and M-Mode Measurements

The recent development of 3-dimensional (3D) surface detection algorithm of the endocardial and epicardial surfaces from real-time 3D echocardiographic (RT3DE) datasets allows direct semiautomated quantification of left ventricular mass (LVM). Our aims were to (1) evaluate the accuracy of RT3DE measurements of LVM using this algorithm against cardiac magnetic resonance (CMR) reference and (2) compare RT3DE LVM with conventional M-mode, 2-dimensional (2D), and RT3DE-guided biplane measurements. A total of 205 patients were studied in 2 protocols: (1) RT3DE and CMR imaging was performed on the same day in 55 subjects; (2) in an additional 150 subjects, RT3DE, 2D, and M-mode images were acquired. In both protocols, RT3DE endocardial and epicardial surfaces were semiautomatically identified at end diastole (QLab, Philips Medical Systems, Andover, MA) to calculate LVM. CMR, 2D, and M-mode-derived LVM were obtained using standard techniques. A significant correlation (r = 0.95) was noted between RT3DE and CMR-derived LVM with a small bias of -2 g. M-mode-derived LVM measurements (175 +/- 64 g) were significantly larger than RT3DE LVM (123 +/- 39 g, bias: 52 g) with moderate correlation (r = 0.76). No significant differences in LVM were noted between 2D (125 +/- 42 g) and RT3DE values (bias: 1.2 g) with good correlation (r = 0.91, P < .001). However, the best correlation was noted between RT3DE and RT3DE-guided biplane LVM values (r = 0.95, P < .001, bias: -4.6 g). Intraobserver, interobserver variability, and test-retest variability of the RT3DE measurements were 9%, 12%, and 6%, respectively. RT3DE imaging using the 3D surface detection algorithm allows accurate and reproducible measurements of LVM. RT3DE-guided biplane technique can be used as an accurate time-saving alternative in clinical practice.

MSCT labelling for pre-operative planning in cardiac resynchronization therapy

The objective of this paper is twofold: (i) to show how multislice computed tomography (MSCT) data sets bring the information required for cardiac resynchronisation therapy (CRT) planning; (ii) to demonstrate the feasibility of 3D navigation into the veins where left ventricular leads have to be placed. The former has been achieved by exploring and labelling the cardiac structures of concern, the latter has been performed by using the concept of virtual navigation with high resolution surface detection and estimation algorithms.

Time reversal focusing through a human skull optimized by skull surface detection algorithm

High-intensity focused ultrasounds are mainly envisioned to cure liver and prostate cancer. In order to cure brain tumors, it has been shown that time reversal coupled with amplitude compensation is an interesting technique to correct the aberrations induced by a human skull. This focusing technique should be improved by compensating the amplitude at the exact location of the skull. To measure the geometrical shape of the skull, two different techniques are investigated and compared in terms of precision and duration of the measurement. The first method is based on the measurement of the propagation time from the array to the skull and back of pulsed signals emitted and received by each pair of transducers of an echographic array of 128 transducers. For each pair, the reflection point on the skull is located on an ellipse whose centers correspond to the positions of the emitting and receiving elements. The envelope of the complete set of ellipses is then extracted, giving the shape of the skull. This technique is compared to echographic imaging of the skull using the same array. Comparisons between these techniques will be presented using the Phased Array Simulation software and the corresponding experimental focusing.

The effect of phantom wall thickness on volume determination in SPET

Assessment, using phantoms, of a a three-dimensional, second-derivative, surface-detection algorithm for accurately determining the volumes of structures in single photon emission tomography (SPET) has shown the performance of the algorithm to be highly dependent on phantom wall thickness. With a 30% background activity level, the volume of a 600-cm3 cylinder with wall thickness approximately 0.4 mm was measured to an accuracy of approximately 1%, whereas for an almost identical cylinder with a wall thickness of approximately 3 mm the measured volume was underestimated by about 14%. To further investigate this 'wall-thickness' effect, theoretically generated SPET data have been produced, simulating a set of low-contrast cylindrical phantoms with identical internal dimensions and wall thicknesses ranging from 0 to 6 mm. These image data have demonstrated a 10% reduction in the calculated volume for wall thicknesses as thin as 1 mm. A less acute dependence is also demonstrated for a threshold-based quantitation algorithm, where a wall thickness of 4-5 mm is required to produce an effect of similar magnitude. The underlying cause behind this 'wall-thickness' effect is undoubtedly the perturbation in the count profile across the surface boundary, which results form the cold region of the phantom wall. Thus, phantom wall thickness will have an effect on the performance of most automated quantitation techniques, both two- and three-dimensional, since the majority must incorporate some form of analysis on this count profile.

Fast surface tracking in three-dimensional binary images

In medical 3-dimensional display, an input scene is represented by an array of volume elements (abbreviated as “voxels”), and an object in the scene is specified as a “connected” set of voxels. In such applications, surface tracking is an important, and often time-consuming, precursory step. One of the most efficient surface detection algorithms reported in the literature tracks a specified surface of a 3-dimensional object by visiting each boundary face in the surface twice. In this paper, we present a new definition of discrete objects and boundaries that leads to a modified algorithm, which, on the average, visits only one-third of the boundary faces twice (and the rest once). The algorithm has been implemented in our display software package and is found to achieve a run-time reduction of approximately 35%. This timing includes the computation of surface-normal information which is needed for the realistic rendering of surfaces. Without this computation, the saving would be about 55%.

The theory, design, implementation and evaluation of a three-dimensional surface detection algorithm

In many three-dimensional imaging applications the three-dimensional scene is represented by a three-dimensional array of volume elements, or voxels for short. A subset Q of the voxels is specified by some property. The objects in the scene are then defined as subsets of Q formed by voxels which are “connected” in some appropriate sense. It is often of interest to detect and display the surface of an object in the scene, specified, say, by one of the voxels in it. In this paper, the problem of surface detection is translated into a problem of traversal of a directed graph, G. The nodes of G correspond to faces separating voxels in Q from voxels not in Q. It is proven that connected subgraphs of G correspond to surfaces of connected components of Q (i.e., of objects in the scene). Further properties of the directed graph are established which allow us to keep the number of marked nodes (needed to avoid loops in the graph traversal) to a small fraction of the total number of visited nodes. This boundary detection algorithm has been implemented. We discuss the interaction between the underlying mathematical theory and the design of the working software. We illustrate the software by some clinical studies in which the input is computed tomographic (CT) data and the output is dynamically rotating three-dimensional displays of isolated organs. Even though the medical application leads to very large-scale problems, our theory and design allows us to use our method routinely on the minicomputer of a CT scanner.

The theory, design, implementation and evaluation of a three-dimensional surface detection algorithm

In many three-dimensional imaging applications the three-dimensional scene is represented by a three-dimensional array of volume elements, or voxels for short. A subset Q of the voxels is specified by some property. The objects in the scene are then defined as subsets of Q formed by voxels which are “connected” in some appropriate sense. It is often of interest to detect and display the surface of an object in the scene, specified say by one of the voxels in it. In this paper, the problem of surface detection is translated into a problem of traversal of a directed graph, G. The nodes of G correspond to faces separating voxels in Q from voxels not in Q. It has been proven that connected subgraphs of G correspond to surfaces of connected components of Q (i.e., of objects in the scene). Further properties of the directed graph have been proven, which allow us to keep the number of marked nodes (needed to avoid loops in the graph traversal) to a small fraction of the total number of visited nodes. This boundary detection algorithm has been implemented. We discuss the interaction between the underlying mathematical theory and the design of the working software. We illustrate the software on some clinical studies in which the input is computed tomographic (CT) data and the output is dynamically rotating three-dimensional displays of isolated organs. Even though the medical application leads to very large scale problems, our theory and design allows us to use our method routinely on the minicomputer of a CT scanner.