Car tracking by quantised input LMS, QX-LMS algorithm in traffic scenes

The tracking algorithm is an important tool for motion analysis in computer vision. A new car tracking algorithm is proposed which is based on a new clipping technique in the field of adaptive filter algorithms. The uncertainty and occlusion of vehicles increase the noise in vehicle tracking in a traffic scene, so the new clipping technique can control noise in prediction of vehicle positions. The authors present a new quantised version of the LMS, namely the QX-LMS algorithm, which has a better tracking capability in comparison with the clipped LMS (CLMS) and the LMS and also involves less computation. The threshold parameter of the QX-LMS algorithm causes controllability and the increase of tracking and convergence properties, whereas the CLMS and LMS algorithms do not have these capabilities. The QX-LMS algorithm is used for estimation of a noisy chirp signal, for system identification and in car tracking applications. Simulation results for noisy chirp signal detection show that this algorithm yields a considerable error reduction in comparison to the LMS and CLMS algorithms. The proposed algorithm, in tracking some 77 vehicles in different traffic scenes, shows a reduction of the tracking error relative to the LMS and CLMS algorithms.     

Curve fitting space for classification

In this paper, a curve fitting space (CFS) is presented to map non-linearly separable data to linearly separable ones. A linear or quadratic transformation maps data into a new space for better classification, if the transformation method is properly guessed. This new CFS space can be of high or low dimensionality, and the number of dimensions is generally low, and it is equal to the number of classes. The CFS method is based on fitting a hyperplane or curve to the learning data or enclosing them into a hypersurface. In the proposed method, the hyperplanes, curves, or cortex become the axis of the new space. In the new space, a linear support vector machine multi-class classifier is applied to classify the learn data.     

Fuzzy cost support vector regression on the fuzzy samples

This paper presents a new version of support vector regression (SVR) named Fuzzy Cost SVR (FCSVR) with a unique property of operating on fuzzy data where fuzzy cost (fuzzy margin and fuzzy penalty) are maximized. This idea admits to have uncertainty in the penalty and margin terms jointly. Robustness against noise is shown to be superior in the experimental results as a property compared with conventional SVR.     

Sediment Flushing of Reservoirs under Environmental Considerations

Freshwater reservoirs are under threat because of huge amounts of sediments deposited annually. Sediment flushing seems to be effective to preserve reservoir storage, but it may have negative environmental impacts on downstream ecosystems such as fish mortality. Therefore, providing a suitable flushing strategy that could be compatible with the river ecosystem downstream is of great importance. Two numerical models were developed in this paper to predict the suspended sediment concentration (SSC) on the reservoir-river system and effects of different flushing scenarios on aquatic life. Developed models were applied to the Dez Resevoir system in the southwest of Iran which has suffered from the sediment problems in two last decades. The suitable values for flushing time, concentration limits, and flushing discharge have been recommended in this research by use of the existing information and previous flushing records, as well as field measurement and modeling. Based on social, environmental and technical limitations, March is the appropriate time for flushing. After hydraulic simulation of different flushing scenarios and sediment routing along the river, flushing with 1275 and 800 cubic meter per second with 30 and 20 g per lit concentration in dry and wet season respectively are feasible and have minimum environmental impacts.     

Effect of pore geometry and loading direction on deformation mechanism of rapid prototyped scaffolds

Rapid prototyping is a promising technique for producing tissue engineering scaffolds due to its capacity to generate predetermined forms and structures featuring distinct pore architectures. The objective of this study is to investigate the influences of different pore geometries and their orientation with respect to the compressive loading direction on mechanical responses of scaffolds. Plastic models of scaffolds with cubic and hexagonal unit cells were fabricated by three-dimensional (3-D) printing. An in situ imaging technique was utilized to study the progressive compressive deformation of the scaffold models. In both cubic and hexagonal geometries, organized buckling patterns relevant to each unit cell were observed at the onset of plastic deformation. These patterns were in good agreement with the elastic stress concentration patterns generated by finite element simulation. Uniaxial compression tests conducted on both geometries revealed that the stress–strain pattern may vary significantly by changing the loading direction. A secondary stress softening phenomenon was observed where limited pore deformation caused 3-D structure bending. However, when loading direction was adjusted such that deformation was localized on all pores simultaneously, a monotonically increasing stress was observed. These results accentuate the critical role of pore geometry and orientation with respect to the principal loading axis in designing functional tissue engineering scaffolds.     

Development of endometriosis from embryonic duct remnants

Closed rupture of the anterior tibial tendon is rare. Fewer than 50 cases have been reported in the literature, perhaps because the symptoms are often neglected by the patient as well as by the doctor. Most often the rupture occurs as a consequence of a sudden plantar flexion of the ankle and pre-existing degenerative changes of the tendon tissue due to systemic disease or iatrogenic local corticoid injections. Clinically, the tendon rupture presents as acute weakness of the ankle extensors without reddening, swelling or neurological signs. Differential diagnoses such as anterior tibial syndrome and peroneal nerve palsy can thus be excluded. During the clinical examination the distal stump and the discontinuity of the anterior tibial tendon are often palpable. The clinical diagnosis can be confirmed by an ultrasound examination. An operation may be undertaken up to 3 months after the injury. Preferred procedures are end-to-end anastomosis and transosseous refixation of the tendon, followed by a plaster cast for 6 weeks post-operatively. High-risk patients can be treated conservatively, but the functional results are less satisfactory. We describe a case of an acute closed rupture of the anterior tibial tendon. A yet unpublished method of osseous reinsertion of the tendon is presented.     

3D printing: an emerging tool for novel microfluidics and lab-on-a-chip applications

In the past few years, 3D printing technology has witnessed an explosive growth, penetrating various aspects of our lives. Current best-in-class 3D printers can fabricate micrometer scale objects, which has made fabrication of microfluidic devices possible. The highest achievable resolution is already at nanometer scale, which is continuing to drop. Since geometric complexity is not a concern for 3D printing, novel 3D microfluidics and lab-on-a-chip systems that are otherwise impossible to produce with traditional 2D microfabrication technology have started to emerge in recent years. In this review, we first introduce the basics of 3D printing technology for the microfluidic community and then summarize its emerging applications in creating novel microfluidic devices. We foresee widespread utilization of 3D printing for future developments in microfluidic engineering and lab-on-a-chip technology.     

Flood flow forecasting using ANN,ANFIS and regression models

Flood prediction is an important for the design, planning and management of water resources systems. This study presents the use of artificial neural networks (ANN), adaptive neuro-fuzzy inference systems (ANFIS), multiple linear regression (MLR) and multiple nonlinear regression (MNLR) for forecasting maximum daily flow at the outlet of the Khosrow Shirin watershed, located in the Fars Province of Iran. Precipitation data from four meteorological stations were used to develop a multilayer perceptron topology model. Input vectors for simulations included the original precipitation data, an area-weighted average precipitation and antecedent flows with one- and two-day time lags. Performances of the models were evaluated with the RMSE and the R ². The results showed that the area-weighted precipitation as an input to ANNs and MNLR and the spatially distributed precipitation input to ANFIS and MLR lead to more accurate predictions (e.g., in ANNs up to 2.0 m³ s^?1 reduction in RMSE). Overall, the MNLR was shown to be superior (R ² = 0.81 and RMSE = 0.145 m³ s^?1) to ANNs, ANFIS and MLR for prediction of maximum daily flow. Furthermore, models including antecedent flow with one- and two-day time lags significantly improve flow prediction. We conclude that nonlinear regression can be applied as a simple method for predicting the maximum daily flow.