Robust object recognition using a color co-occurrence histogram and the spatial relations of image patches

We propose a robust object recognition system where patch-based pyramid images and the spatial relationships among patches are utilized for our image model. In particular, both a color histogram (CH) and a color co-occurrence histogram (CCH) are applied to obtain image features for each patch. The locations of subregions to be tested are decided by a particle filter in our matching process. We show that the performance of object recognition can be improved by using the spatial relationships among patches. To show the validity of our proposed method, we employ input images from various environments as test images. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

Measurements of children's exposures to particles and nitrogen dioxide in Santiago, Chile

An exposure study of children (aged 10-12 years) living in Santiago, Chile, was conducted. Personal, indoor and outdoor fine and inhalable particulate matter (< 2.5 .m in diameter, PM2.5 and < 10 microm in diameter, PM10, respectively), and nitrogen dioxide (NO2) were measured during pilot (N = 8) and main (N = 20) studies, which were conducted during the winters of 1998 and 1999, respectively. For the main study, personal, indoor and outdoor 24-h samples were collected for five consecutive days. Similar mean personal, indoor and outdoor PM2.5 concentrations (69.5, 68.5 and 68.1 microg/m3, respectively) were found. However, for coarse particles (calculated as the difference between measured PM10 and PM2.5, PM2.5-10), indoor and outdoor levels (35.4 and 47.4 microg/m3) were lower than their corresponding personal exposures (76.3 microg/m3). Indoor and outdoor NO2 concentrations were comparable (35.8 and 36.9 ppb) and higher than personal exposures (25.9 ppb). Very low ambient indoor and personal O3 levels were found, which were mostly below the method's limit of detection (LOD). Outdoor particles contributed significantly to indoor concentrations, with effective penetration efficiencies of 0.61 and 0.30 for PM2.5 and PM2.5-10, respectively. Personal exposures were strongly associated with indoor and outdoor concentrations for PM2.5, but weakly associated for PM2.5-10. For NO2, weak associations were obtained for indoor-outdoor and personal-outdoor relationships. This is probably a result of the presence of gas cooking stoves in all the homes. Median I/O, P/I and P/O ratios for PM2.5 were close to unity, and for NO2 they ranged between 0.64 and 0.95. These ratios were probably due to high ambient PM2.5 and NO2 levels in Santiago, which diminished the relative contribution of indoor sources and subjects' activities to indoor and personal PM2.5 and NO2 levels.     

Effect of cooling rate during hot stamping on low cyclic fatigue of boron steel sheet

Boron steel is widely used throughout the automobile industry due to its high tensile strength and hardenability. When boron steel is used for body parts, only high strength is required for crashworthiness. However, when boron steel is used for chassis parts, a high fatigue life is needed. The microstructure of boron steel is mainly affected by the cooling rate during hot stamping. Therefore, this study investigated the low cyclic fatigue life according to the cooling rate. The fatigue life increased at a low strain amplitude when the cooling rate was fast. However, at a high strain amplitude, the fatigue life decreased, due to the low ductility and fracture toughness of the martensite formed by rapid cooling. Martensite formed by a fast cooling rate shows excellent fatigue life at a low total strain amplitude; however, a multiphase microstructure formed by a slow cooling rate is recommended if the parts experience high and low total strain amplitudes alternately. In addition, the cooling rate has little effect on the distribution of solute boron and boron precipitations, so it is expected that boron rarely affects low cyclic fatigue.     

Indentation size effect and shear transformation zone size in a bulk metallic glass in two different structural states

The existence of an indentation size effect (ISE) in the onset of yield in a Zr-based bulk metallic glass (BMG) is investigated by employing spherical-tip nanoindentation experiments. Statistically significant data on the load at which the first pop-in in the displacement occurs were obtained for three different tip radii and in two different structural states (as-cast and structurally relaxed) of the BMG. Hertzian contact mechanics were employed to convert the pop-in loads to the maximum shear stress underneath the indenter. Results establish the existence of an ISE in the BMG of both structural states, with shear yield stress increasing with decreasing tip radius. Structural relaxation was found to increase the yield stress and decrease the variability in the data, indicating “structural homogenization” with annealing. Statistical analysis of the data was employed to estimate the shear transformation zone (STZ) size. Results of this analysis indicate an STZ size of ～25 atoms, which increases to ～34 atoms upon annealing. These observations are discussed in terms of internal structure changes that occur during structural relaxation and their interaction with the stressed volumes in spherical indentation of a metallic glass.     

A review on the analysis and experiment of fluid flow and mixing in micro-channels

The studies with respect to micro-channels and micro-mixers are expanding in many dimensions. Most significant area of micro-mixer study is the flow analysis in various micro-channel configurations. The flow phenomena in microchannel devices are quite different from that of the macro-scale devices. An attempt is made here to review the important recent literature available in the area of micro-channel flow analysis and mixing. The topics covered include the physics of flow in micro-channels and integrated simulation of the micro-channel flow. Also, the flow control models and electro-kinetically driven micro-channel flows are dealt in detail. A survey of important numerical methods, which are currently popular for micro-channel flow analysis, is carried out. Different options for mixing in microchannels are provided, in sufficient detail.     

A study on the development of a web based urban transit reliability evaluation system

Over the past twenty years, the maintenance system developed, and its importance increased. For the effective maintenance, the maintenance system has been developed by introducing the concept of RCM (Reliability Centered Maintenance). This research is to develop an effective maintenance system for urban transit based on the concept of RCM. RCM is a systematic approach to develop a cost-effective maintenance strategy based on the various components’ reliability of the subject system. The final process of RCM determines appropriate failure maintenance strategies. For realization of RCM, reliability evaluation framework has been studied to compute the reliability index for urban transit. The framework requires the following processes: BOM (Bill of Materials), RBD (Reliability Block Diagram) based on FBD (Function Block Diagram), and failure code classification. The goal of this paper is to define the maintenance procedure for the subject system since successful maintenance system depends on an automated maintenance plan. This plan can be scheduled effectively by collecting and analyzing data from maintenance experience. For doing this, this paper proposes the web-based maintenance system for collecting data and the computing of MTBF (Mean Time between Failures) at the maintenance stage for analyzing data.     

A study on the optimum design of high-speed low-floor bogie with independently rotating wheels

The low-floor bogie is a prior technology in countries and companies that want to develop the tram. The Low-floor tram (LFT), which includes low-floor bogies, is easy to embark and disembark because of the low floor height. In addition, it can be driven on urban as well as rural tracks. Furthermore, emissions such as NOx and SOx can be reduced. Due to these advantages, this innovative technology is expected to change the public transport system. To improve utilization in a downtown area, the technology for the low-floor bogie should satisfy conditions of a high-speed of over 80 km/h and minimum radius within a 25 mR curve for smooth running on a track that has a severe turning radius. Moreover, the wheelset should not be located in the bogie, and the components inside the bogie need to be wellarranged to satisfy the full low-floor condition. In this study, to develop an over-80 km/h class high-speed low-floor bogie that can be driven safely on a 25 mR curved track, a conceptual design of the LFT multibody dynamics model was constructed and dynamic characteristics were assessed by dynamic analysis. The modeling modification with Independently rotating wheels (IRW) needed to steer actively through semi-active suspension and the optimization using Design of experiments (DOE) were then performed. Through DOE method, the optimum combination of design parameters could be obtained and, the driving performances such as ride stability, comfort and safety of the LFT could then be improved about 7 %. The results of this work are available to detail design and development of LFT.     

Module-based Failure Propagation (MFP) model for FMEA

Failure mode and effect analysis (FMEA) is one of the most important activities in the quality management of product design. Within the field of product development domain, FMEA is regarded as a key driver for product quality improvement and cost reduction. There have been various FMEA approaches considering several aspects such as knowledge management, criticality analysis, and product structure or function model to conduct more effective and systematic FMEA. But previous approaches are generally based on an ad hoc model so that they have limitations such as insufficient identification on failures, low reusability of previous FMEAs, poor understanding on failure information, high dependency of expert knowledge on failure propagation, and so on. To overcome these limitations, this study proposes a formalized function model for FMEA, which is named module-based failure propagation (MFP) model. The MFP model is proposed based on the function behavior state scheme. The MFP model consists of function decomposition tree model, configuration flow graph model, function rule, and failure rule. This study describes how to build the MFP model and introduces how to carry out FMEA with the proposed MFP model. To show the benefit of the proposed MFP model, a FMEA case study on a car air purifier is performed.     

Effect of welding heat input on fatigue life of quenched boron steel and FB steel lap joint

The effect of welding heat input on the fatigue life of a quenched boron steel and ferrite-bainite (FB) steel lap joint was investigated. Boron steel was quenched and welded with FB steel in heat input ranging from 0.29 to 0.67 kJ/mm. Boron, which can increase hardenability, affected the microstructure and hardness of the weld metal and heat affected zone (HAZ). The hardness of the weld metal and HAZ increased with decreasing welding heat input, and the high hardness of the weld metal and boron steel HAZ prevented the initiation of cracks in the stress concentration area around the bead. The bead width increased with increasing heat input, and the results of finite element method (FEM) showed that the maximum stress in the notch of the weld joint decreased when the bead width was increased. That is to say, the fatigue life increased when the weld joint had wider bead width. Finally, while the fatigue life was affected by the residual stress, the variation of the welding heat input used in this study had hardly any affect on the residual stress distribution.     

Study on mechanical properties of austenitic stainless steel depending on heat input at laser welding

The austenitic stainless steels used in various industrial fields require low heat input for welding. Laser welding is an excellent welding method in this respect. This study investigates the effect of laser welding speed on the mechanical properties at welding speed of 1.0, 1.5 and 2.0 m/min, using STS304L of austenitic stainless steel. The microstructures of fusion zone (FZ) show a two phase structure consisted of austenite and δ-ferrite, and δ-ferrite in the fusion zone tend to decrease with increase of welding speed. Meanwhile, the mechanical properties were excellent at the welding speed of 1.5 m/min for tensile, bending and impact tests.