BACKGROUND AND OBJECTIVE:Accurate object segmentation in medical images is a crucial step in medical diagnosis and other applications. Despite years of research on automatic segmentation approaches, achieving clinically acceptable image quality remains challenging. Interactive segmentation is seen as a promising alternative; thus, we propose a new interactive segmentation framework based on a progressive workflow to reduce user effort and provide high-quality results.METHOD:First, our approach encodes user-provided region clicks and edge scribbles using our proposed disk and curve transform. Then, it is followed by refinement with a transformer-based module that extracts effective features from the outputs of the convolutional neural network (CNN) and the extra input maps.RESULT:Extensive experiments conducted on various medical images, including ultrasound (US), computerized tomography (CT), and magnetic resonance images (MRI), have demonstrated the effectiveness of our new approach over the state-of-the-art alternatives.CONCLUSION:The proposed framework can achieve high-quality segmentation using minimal interactions without the substantial cost of manual segmentation.

Severe wear is a common damage mechanism in railway turnouts, which strongly affects the dynamic performance of railway vehicles and maintenance costs of tracks. This article explores the effects of profile wear on contact behaviors in the wheel–rail/switch contact and dynamic interaction, and nominal and measured worn turnout rail profiles are used as boundary conditions of wheel–rail contact. The calculation of the dynamic loads and the resultant contact stresses and internal stresses makes it possible to rationally design railway turnouts and correctly select the material to be applied for their components. For these reasons, the multi-body system SIMPACK and finite element software ANSYS are used to calculate the features of load and subsequently distributions of contact stresses and internal stresses in the regions of wheel–turnout components. The results show that profile wear disturbs the distribution of wheel–rail contact point pairs, changes the positions of wheel–rail contact points along the longitudinal direction, and affects the dynamic interaction of vehicle and turnout. For the measured profile in this article, profile wear aggravates vertical dynamic responses significantly but improves lateral dynamic responses. Profile wear disturbs the normal contact situations between the wheel and switch rail and worsens the stress state of the switch rail.

In this paper, a defect detecting system of wheel tread based on linear charge coupled device (CCD) is established, and the image recognition technology of wheel tread defects under this system is studied. Firstly, images of the whole wheel tread are captured through the system. Secondly, median filtering and image contrast enhancement are processed on the defect images, and then the canny edge detection based on gradient histogram is performed. Thirdly, the images after edge detection are used for eight-neighbor boundary tracking, in order to obtain the sizes in-pixel of defects. Finally, the linear CCD camera is calibrated by Zhang checkerboard-like calibration method, and the actual sizes of defects are calculated. The results of this experiment show that the detecting system can identify the wheel tread defects effectively, and obtain the defects' actual width and height accurately through image processing algorithm.

The quality of wheel is especially important for the safety of high speed railway. In this paper, a new ultrasonic array inspection method, the Full Matrix Capture (FMC) has been studied and applied to the high speed railway wheel inspection, especially in the wheel web from the tread. Firstly, the principle of FMC and TFM algorithm is discussed, and then the new optimization is applied to the standard FMC; Secondly the fundamentals of optimization is described in detail and the performance is analyzed. Finally, the experiment has been built with a standard phased array block and railway wheel, and then the testing results are discussed and analyzed. It is demonstrated that this change for the ultrasonic data acquisition and image reconstruction has higher efficiency and lower cost comparing to the FMC's procedure.

The mainstream methods of pantograph slide edge detection are based on canny operator and multi-scale wavelet. The former has good single edge response but the edge is fractured, the latter performs good edge continuity but contains excessive edge points. This paper combines the advantages of both methods and proposes as an improved multi-scale wavelet edge detection method based on canny criteria. Firstly we filtered the pantograph image with edge-preserving symmetric near neighbor filter. Secondly calculated the Gaussian wavelet modulus and arguments at all levels of scale, then suppressed the non-maxima value of modulus along the corresponding arguments. At last, we integrated the modulus drawings at all levels of scale, and connected edge with applicable dual-threshold. Experiments results show that the improved algorithm has both satisfactory performances in single edge response and edge continuity, it markedly improves the efficiency of edge detection algorithm. Peak signal to noise ratio (PSNR) analysis finds that the improved algorithm exceeds canny operator and traditional multi-scale wavelet edge detection. Moreover, it has higher positioning accuracy, clearer details and better noise performance.

In this paper, a design concept of sparse array is applied to total focusing imaging technique. During the process, the elements of a phased array are partially selected according to the theory of effective aperture, in order to reduce the amount of captured data and shorten the processing time of TFM imaging. Simulated imaging with Field II program and experimental verification on two standard phased array specimens are performed, based on three selected arrays which are sparsely as 1/2, 1/4 and 1/8. The results compared with the full array in terms of image quality and computing efficiency, show that the computing efficiency of TFM imaging can get a rise of 40 percent by adopting the sparse array design. Meanwhile, an approximate resolution can be achieved at the lowest array sparse rate of 1/4. Affected by the side-lobes of the ultrasonic beam, contrast of the image may suffer little loss.

To solve the stability problem of continuous welded rail (short for CWR) on bridge in the small radius curve section, a sleeper connector structure was proposed, which is used to connect sleepers together and strengthen the stability of CWR. A FEM model was established to study the effect of the CWR stability strengthening scheme. This study result shows that the integrity of the track structure and the stability of CWR in the small radius curve section of which the radius is from 250m to 600m will be both improved by the strengthening scheme, and the larger the curve radius is, the better the CWR stability improvement will be. In order to guarantee the strength of sleeper connector and connecting bolt, it is suggested that the sleeper connectors at beam gaps should not be connected, and small resistance fasteners should be used in strengthening area.

The past rail wear measurement methods and equipments are summarized and the development process of portable inspection car for measuring rail wear is presented. The measurement system of this portable inspection car is developed based on the software platform of Labview. The car structure design is delicate and matched well with the image collecting device. This portable inspection car can measure the rail wear every 2.5 cm along the railway line and the measuring accuracy is 0.1mm.

Train wheels should undergo heat treatment, thus producing circumferential compressive force on the tread to void crack propagation, and this way eventually increases the train wheel service life, excluding situations like wheel burst due to crack. Hem shoes braking will lead to calorification and cooling of wheel rim, in this case stress state of wheel rim will would change under the influence of caloric and mechanical load, and compressive stress will turn to circumferential tensile force. For wheels in service, small flaws appear in the shallow layer of wheel tread, and will expand into depth. Consequently, detecting of wheel rim stress state should be attached more importance. Ultrasonic testing technology is an extremely important method for convenient, fast and non-destructive detecting in routine application of wheel sets. Verification of ultrasonic detecting method in detecting stress of wheel rim is given in this paper. On experimentation platform established with wheel rim stress detecting equipment and tensile stress mechanism, ultrasonic wheel rim stress detecting method is verified to be reliable and stable in the light of contrast between theoretical and experimental results. Basing on wheel rim stress measurement method using transverse straight probe, results are obtained by measuring test block and wheel rim stress, and ultrasonic wheel rim stress detecting method is approved to be feasible in application.

In the phased array imaging system, in order to get better images,the lateral resolution need to be improved and the grating lobes need to be eliminated. Restrainning the width of the main lobe is a way to improve the lateral resolution of the system. This article designed the non-periodic distribution of elements to decrease the width of main lobe of the transducer using sparse matrix methods. In this way, we can improve the lateral resolution, resolve the problem of high-density of elements in phased array element, eliminate the impact of the gate lobe, lowerring the effects of radiation in the neighboring elements, save costs and overcome the difficulty in production of elements.

Aimed at the problem of ultrasonic axis's sound pressure impacting on flat-bottom hole defect echo wave,this paper studies the approach of ultrasonic flaw based equivalent method.Firstly,according to the sound pressure model from the point sound source to a point on the sound axis,the transmitting sound field of wave axis has been set up and echo sound pressure model of the flat-bottom hole reflector has been established with Fresnel approximation theory.Then the effects of the dimension of flat-bottom hole,the depth of flat-bottom hole,the sound source size and the transmission frequency on the echo wave of the flat-bottom hole have been analyzed theoretically and simulated with experiment.The results showed that the model could be a guide to the quantitative nondestructive assessment.

Speaker recognition is to recognize speaker's identity from its voice which contains physiological and behavioral characteristics unique to each individual. In this paper, the artificial neural network model, which has very good capacity of non-linear division in characteristic space, is used for pattern matching. The speaker's sample characteristic domain is built for his mixed voice characteristic signals based on Kmeanlbg algorithm. Then the dimension of the inputting eigenvector is reduced, and the redundant information is got rid of. On this basis, BP neural network is used to divide capacity area for characteristic space nonlinearly, and the BP neural network acts as a classifier for the speaker. Finally, a speaker recognition system based on the neural network is realized and the experiment results validate the recognition performance and robustness of the system.

Double slip turnout is common in heavy haul railway and due to its complex structure, its maintenance is difficult as well as important. Focusing on No.12 double slip turnout (75 kg/m rail), this study analyzed the intensity of the different slotted rubber pad under iron plate and track stiffness of turnout zone based on the finite element method with a purpose of improving the performance of the double slip turnout. Results show that: the depth of groove on rubber pad and groove center distance have great influences on the equivalent stresses of iron plate and the pad; under pressure of 85.52 kN on sleeper, the maximal equivalent stresses of iron plate and the pad under iron plate are 72.959 and 2.141 MPa, respectively, both smaller than their allowable value; the maximal longitudinal variation rates of entire track stiffness are 1.42 and 1.34 respectively; the stiffness uniformity of the diamond-shaped double slip turnout can meet the requirement that the train runs at or below 120 km/h. All these achievements can be used as the theoretical guidance for performance optimization of double slip turnout and improving the running stability of heavy haul trains as well as extending service life of the turnout. © Maxwell Scientific Organization, 2012.

The irregularity is a key factor affecting the wheel-rail contact geometry relationship. In this paper, we calculated the wheel-rail contact points at typical sections and obtained the longitudinal variation of the wheel-rail geometry relationship with the trace line method. The profile of the key rail sections was matched by cubic spline curve, and the shape interpolation was realized in non-controlling sections. The results show that the roll angles at each typical section increases gradually with the enlargement of track alignment irregularity. When the flange contact occurs, the roll angle increases dramatically. Proper track alignment irregularity towards the switch rail improves the structure irregularity of the turnout.

In this paper, an Ultrasonic testing method for inservice wheel by use of phased array technology is introduced. To meet the ultrasonic test criterion of inservice CRH wheel, groups of phased array probes and conventional probes are combined used. By study echo wave in wheel rim and wheel disk form different probes, defect analyzing methods in wheel rim and wheel disk are studied. Tests on CRH reference wheels with manmade defects and on CRH wheels online shows that the arrangement of probes and defect analyzing methods can meet the requirement of field application.

Based on radiant sound field expression for single rectangular piezoelectric wafer,this paper gives analysis the infection each factor causing to sound pressure intensity and directivity,concrete methods to minimize major lobe while narrowing side lobes is put forward.As an example,the paper derives superimposed sound field formula for 16-wafer phased array probe,analyzes relationship of superimposed sound pressure and distance between focus and transducer,simulates and discusses influence on the quality of synthetic sound beam of phased array probe by factors such as element number,distance of adjacent elements.Simulation result is considerably consistent to related theory,which provides reference for evaluation and measurement for sound distribution and production of high-quality phased array probes.

In this paper, a new automatic defect localization method for mobile wheel ultrasonic image from a mobile wheel inspection system. This wheel inspection system adopting the phased array ultrasonic technology is widely used in railway industries for detecting all kinds of wheel defects in the wheel rim and disk area. However, the echo wave in the railway wheel is highly complex because of the unregularly wheel geometry, especially with the bore hole in the wheel disk, so that the common evaluation method by the A-Scan and B-scan image fails. The objective of our study is to develop a data processing algorithm for the B-scan image of wheel, to identify the defect echo wave image and assist the processing of the wheel quality evaluation. The research work described in the paper has thus been done with this objective in mind and it concerns a method that we have developed for automatic detection of defects. This method with the morphology processing and region growing can realize the automatic locate the defects image and increase the possibility of defects (POD) in the wheel.