Criteria

Text:
Display:

Results

Viewing 1 to 30 of 366
2017-09-23
Technical Paper
2017-01-1971
Sihan Chen, Libo Huang, Xin Bi, Jie Bai
Abstract For sensing system, the trustworthiness of the variant sensors is the crucial point when dealing with advanced driving assistant system application. In this paper, an approach to a hybrid camera-radar application of vehicle tracking is presented, able to meet the requirement of such demand. Most of the time, different types of commercial sensors available nowadays specialize in different situations, such as the ability of offering a wealth of detailed information about the scene for the camera or the powerful resistance to the severe weather for the millimeter-wave (MMW) radar. The detection and tracking in different sensors are usually independent. Thus, the work here that combines the variant information provided by different sensors is indispensable and worthwhile. For the real-time requirement of merging the measurement of automotive MMW radar in high speed, this paper first proposes a fast vehicle tracking algorithm based on image perceptual hash encoding.
2017-09-23
Journal Article
2017-01-1969
Yuanxin Zhong, Sijia Wang, Shichao Xie, Zhong Cao, Kun Jiang, Diange Yang
Abstract Real-time reconstruction of 3D environment attributed with semantic information is significant for a variety of applications, such as obstacle detection, traffic scene comprehension and autonomous navigation. The current approaches to achieve it are mainly using stereo vision, Structure from Motion (SfM) or mobile LiDAR sensors. Each of these approaches has its own limitation, stereo vision has high computational cost, SfM needs accurate calibration between a sequences of images, and the onboard LiDAR sensor can only provide sparse points without color information. This paper describes a novel method for traffic scene semantic segmentation by combining sparse LiDAR point cloud (e.g. from Velodyne scans), with monocular color image. The key novelty of the method is the semantic coupling of stereoscopic point cloud with color lattice from camera image labelled through a Convolutional Neural Network (CNN).
2017-09-23
Technical Paper
2017-01-1975
Wenhui Li, Wenlan Li, Jialun Liu, Yuhao Chen
Abstract Vehicle detection has been a fundamental problem in the research of Intelligent Traffic System (ITS), especially in urban driving environment. Over the past decades, vision-based vehicle detection has got a considerable attention. In addition to the rich appearance information, the stereo vision method also provides depth information, which could achieve higher accuracy and precision. In this paper, a hybrid method for stereo vision-based real-time vehicle detection in urban environment is proposed. Firstly, we extract vehicle features and generate the Region of Interest (ROI). Semi-global Matching (SGM) algorithm is then utilized on the ROIs to generate disparity maps and get the depth information, which could be used to compute the width of each ROI. The noise regions, always with obvious depth variation in the disparity maps are excluded by the clustering approach.
2017-09-23
Technical Paper
2017-01-1977
Xin Bi, Bin Tan, Zhijun Xu, Libo Huang
Abstract Vehicle and pedestrian detection technology is the most important part of advanced driving assistance system (ADAS) and automatic driving. The fusion of millimeter wave radar and camera is an important trend to enhance the environmental perception performance. In this paper, we propose a method of vehicle and pedestrian detection based on millimeter wave radar and camera. Moreover, the proposed method complete the detection of vehicle and pedestrian based on dynamic region generated by the radar data and sliding window. First, the radar target information is mapped to the image by means of coordinate transformation. Then by analyzing the scene, we obtain the sliding windows. Next, the sliding windows are detected by HOG features and SVM classifier in a rough detect. Then using the match function to confirm the target. Finally detecting the windows in a precision detection and merging the detecting windows.
2017-09-23
Technical Paper
2017-01-1978
Yuxiang Feng, Simon Pickering, Edward Chappell, Pejman iravani PhD, Chris Brace
Abstract The major contribution of this paper is to propose a low-cost accurate distance estimation approach. It can potentially be used in driver modelling, accident avoidance and autonomous driving. Based on MATLAB and Python, sensory data from a Continental radar and a monocular dashcam were fused using a Kalman filter. Both sensors were mounted on a Volkswagen Sharan, performing repeated driving on a same route. The established system consists of three components, radar data processing, camera data processing and data fusion using Kalman filter. For radar data processing, raw radar measurements were directly collected from a data logger and analyzed using a Python program. Valid data were extracted and time stamped for further use. Meanwhile, a Nextbase monocular dashcam was used to record corresponding traffic scenarios. In order to measure headway distance from these videos, object depicting the leading vehicle was first located in each frame.
2017-09-23
Technical Paper
2017-01-1997
Cui Hua
Abstract Vision based driving environment perception is current research hotspot in automatic driving field, which has made great progress due to the continuous breakthroughs in the research of deep neural network. As is well known, deep neural network has won tremendous successes in a wide variety of image recognition tasks, such as pedestrian detection and vehicle identification, which have accomplished the commercialization successfully in intelligent monitor system. Nevertheless, driving environment perception has a higher request for the generalization performance of deep neural network, which needs further studies on its design and training methods. In this paper, we presented a new boosted deep neural network in order to improve its generalization performance and meanwhile keep computational budget constant. Above all, the most representative methods to improve the generalization performance of deep neural network were introduced.
2017-09-23
Technical Paper
2017-01-1998
Shun Yang, Weiwen Deng, Zhenyi Liu, Ying Wang
Abstract Intelligent driving, aimed for collision avoidance and self-navigation, is mainly based on environmental sensing via radar, lidar and/or camera. While each of the sensors has its own unique pros and cons, camera is especially good at object detection, recognition and tracking. However, unpredictable environmental illumination can potentially cause misdetection or false detection. To investigate the influence of illumination conditions on detection algorithms, we reproduced various illumination intensities in a photo-realistic virtual world, which leverages recent progress in computer graphics, and verified vehicle detection effect there. In the virtual world, the environmental illumination is controlled precisely from low to high to simulate different illumination conditions in the driving scenarios (with relative luminous intensity from 0.01 to 400). Sedan cars with different colors are modelled in the virtual world and used for detection task.
2017-09-19
Technical Paper
2017-01-2048
Bryan Shambaugh, Patrick Browning
This paper investigates the effect of various magnetic field configurations on an ionized exhaust plume operating under near vacuum conditions. The purpose of this investigation is to determine if deploying a toroidal magnetic field around an ionized exhaust plume can alter the exhaust profile. The test apparatus utilizes a series of twelve N52 grade neodymium magnets mounted on a steel toroid. The design is proposed as a low-cost alternative to toroidal electromagnets. Five different apparatus configurations were tested in this experiment. Each test was documented using 12 sets of photographs taken from a fixed position with respect to the flow. Photographs were taken after the arc jet had run for 10, 20, and 30 seconds. Data from each configuration was compiled using image processing and compared with data from other configurations at corresponding time periods. Two configurations were run as control tests without any magnetic interference.
2017-09-19
Technical Paper
2017-01-2123
Violet Leavers
The need to maintain aircraft in remote, harsh environments poses significant challenges for on-site condition monitoring. For example, in desert assignments or on-board ships, frequent rotation of staff with variable levels of skill requires condition monitoring equipment that is not only robust and portable but also user friendly and requiring a minimum of training to set up and use correctly. The mainstays of any on-site aerospace maintenance program are various fluid and particulate condition monitoring tests that convey information about the current mechanical state of the system. In the front line of these is the collection and analysis of wear debris particles retrieved from a component’s lubricating or power transmission fluid or from magnetic plugs. It is standard practice within the specialist laboratory environment to view and image wear debris using a microscope.
2017-09-04
Journal Article
2017-24-0109
Nic Van Vuuren, Lucio Postrioti, Gabriele Brizi, Federico Picchiotti
Selective Catalytic Reduction (SCR) diesel exhaust aftertreatment systems are virtually indispensable to meet NOx emissions limits worldwide. These systems generate the NH3 reductant by injecting aqueous urea solution (AUS-32/AdBlue®/DEF) into the exhaust for the SCR NOx reduction reactions. Understanding the AUS-32 injector spray performance is critical to proper optimization of the SCR system. Specifically, better knowledge is required of urea sprays under operating conditions including those where fluid temperatures exceed the atmospheric fluid boiling point. Results were previously presented from imaging of an AUS-32 injector spray which showed substantial structural differences in the spray between room temperature fluid conditions, and conditions where the fluid temperature approached and exceeded 104° C and “flash boiling” of the fluid was initiated.
2017-09-04
Technical Paper
2017-24-0126
Christian Zöllner, Dieter Brueggemann
Abstract The removal of particulate matter (PM) from diesel exhaust is necessary to protect the environment and human health. To meet the strict emission standards for diesel engines an additional exhaust aftertreatment system is essential. Diesel particulate filters (DPF) are established devices to remove emitted PM from diesel exhaust. But the deposition and the accumulation of soot in the DPF influence the filter back pressure and therefore the engine performance and the fuel consumption. Thus a periodical regeneration through PM oxidation is necessary. The oxidation behavior should result in an effective regeneration mode that minimizes the fuel penalty and limits the temperature rise while maintaining a high regeneration efficiency. Excessive and fast regenerations have to be avoided as well as uncontrolled oxidations, which may lead to damages of the filter and fuel penalty.
2017-06-05
Technical Paper
2017-01-1874
Tongyang Shi, Yangfan Liu, J Stuart Bolton, Frank Eberhardt, Warner Frazer
Abstract Wideband Acoustical Holography (WBH), which is a monopole-based, equivalent source procedure (J. Hald, “Wideband Acoustical Holography,” INTER-NOISE 2014), has proven to offer accurate noise source visualization results in experiments with a simple noise source: e.g., a loudspeaker (T. Shi, Y. Liu, J.S. Bolton, “The Use of Wideband Holography for Noise Source Visualization”, NOISE-CON 2016). From a previous study, it was found that the advantage of this procedure is the ability to optimize the solution in the case of an under-determined system: i.e., when the number of measurements is much smaller than the number of parameters that must be estimated in the model. In the present work, a diesel engine noise source was measured by using one set of measurements from a thirty-five channel combo-array placed in front of the engine.
2017-06-05
Technical Paper
2017-01-1872
Masao Nagamatsu
Abstract The almost current sound localization methods do not have enough resolution in low frequency sound localization. To overcome this disadvantage, I am now developing the new sound localization method, Double Nearfield Acoustic Holography (DNAH) method. This method is a converted method of conventional Nearfield Acoustic Holography (NAH) method. In this proposing method, the resolution of low frequency sound localization is improved by using sound propagation information on doubled measurement planes. To prove the performance of proposing method, the basic experiments with variable conditions are conducted. In these experiments, the small speakers are used as sound sources. In this paper, to discuss the ability to apply to actual industry, the effect of measurement distance from the sound source is explained. Some experimental results with changing measurement distance are shown in this paper.
2017-05-18
Journal Article
2017-01-9678
G Agawane, Varun Jadon, Venkatesham Balide, R Banerjee
Abstract Liquid sloshing noise from an automotive fuel tank is becoming increasingly important during frequent accelerating/decelerating driving conditions. It is becoming more apparent due to significant decrease in other noise sources in a vehicle, particularly in hybrid vehicles. As a step toward understanding the dynamics of liquid sloshing and noise generation mechanism, an experimental study was performed in a partially filled rectangular tank. A systematic study was performed to understand the effects of critical parameters like fill level and acceleration/deceleration magnitude. Response parameters like dynamic pressure, dynamic force, dynamic acceleration and sound pressure levels along with high speed video images were recorded. The proposed experimental setup was able to demonstrate major events leading to sloshing noise generation. These events in the sloshing mechanism have been analysed from the dynamic sensor data and correlated with high speed video images.
2017-03-28
Technical Paper
2017-01-0753
Marcus Olof Lundgren, Zhenkan Wang, Alexios Matamis, Oivind Andersson, Mattias Richter, Martin Tuner, Marcus Alden, Andersson Arne
Abstract Gasoline partially premixed combustion (PPC) has shown potential in terms of high efficiency with low emissions of oxides of nitrogen (NOx) and soot. Despite these benefits, emissions of unburned hydrocarbons (UHC) and carbon monoxide (CO) are the main shortcomings of the concept. These are caused, among other things, by overlean zones near the injector tip and injector dribble. Previous diesel low temperature combustion (LTC) research has demonstrated post injections to be an effective strategy to mitigate these emissions. The main objective of this work is to investigate the impact of post injections on CO and UHC emissions in a quiescent (non-swirling) combustion system. A blend of primary reference fuels, PRF87, having properties similar to US pump gasoline was used at PPC conditions in a heavy duty optical engine. The start of the main injection was maintained constant. Dwell and mass repartition between the main and post injections were varied to evaluate their effect.
2017-03-28
Technical Paper
2017-01-0109
Yi Zhang, Madeline J. Goh, Vidya Nariyambut Murali
Abstract This work describes a single camera based object distance estimation system. As technology on vehicles is constantly advancing on the road to autonomy, it is critical to know the locations of objects in 3D space for safe behavior of the vehicle. Though significant progress has been made on object detection in 2D sensor space from a single camera, this work additionally estimates the distance to said object without requiring stereo vision or absolute knowledge of vehicle motion. Specifically, our proposed system is comprised of three modules: vision based ego-motion estimation, object-detection, and distance estimation. In particular, we compensate for the vehicle ego-motion by using pin-hole camera model to increase the accuracy of the object distance estimation.
2017-03-28
Technical Paper
2017-01-0099
Jose E. Solomon, Francois Charette
Abstract The proposed technique is a tailored deep neural network (DNN) training approach which uses an iterative process to support the learning of DNNs by targeting their specific misclassification and missed detections. The process begins with a DNN that is trained on freely available annotated image data, which we will refer to as the Base model, where a subset of the categories for the classifier are related to the automotive theater. A small set of video capture files taken from drives with test vehicles are selected, (based on the diversity of scenes, frequency of vehicles, incidental lighting, etc.), and the Base model is used to detect/classify images within the video files. A software application developed specifically for this work then allows for the capture of frames from the video set where the DNN has made misclassifications. The corresponding annotation files for these images are subsequently corrected to eliminate mislabels.
2017-03-28
Technical Paper
2017-01-0395
Xin Xie, Danielle Zeng, Boyang Zhang, Junrui Li, Liping Yan, Lianxiang Yang
Abstract Vehicle front panel is an interior part which has a major impact on the consumers’ experience of the vehicles. To keep a good appearance during long time aging period, most of the front panel is designed as a rough surface. Some types of surface defects on the rough surface can only be observed under the exposure of certain angled sun light. This brings great difficulties in finding surface defects on the production line. This paper introduces a novel polarized laser light based surface quality inspection method for the rough surfaces on the vehicle front panel. By using the novel surface quality inspection system, the surface defects can be detected real-timely even without the exposure under certain angled sun light. The optical fundamentals, theory derivation, experiment setup and testing result are shown in detail in this paper.
2017-03-28
Technical Paper
2017-01-0394
Junrui Li, Ruiyan Yang, Zhen Li, Changqing Du, Dajun Zhou, Lianxiang Yang
Abstract Advanced high-strength steel (AHSS) is gaining popularity in the automotive industry due to its higher final part strength with the better formability compares to the conventional steel. However, the edge fracture occurs during the forming procedure for the pre-strained part. To avoid the edge fracture that happens during the manufacturing, the effect of pre-strain on edge cracking limit needs to be studied. In this paper, digital image correlation (DIC), as an accurate optical method, is adopted for the strain measurement to determining the edge cracking limit. Sets of the wide coupons are pre-strained to obtain the samples at different pre-strain level. The pre-strain of each sample is precisely measured during this procedure using DIC. After pre-straining, the half dog bone samples are cut from these wide coupons. The edge of the notch in the half dog bone samples is created by the punch with 10% clearance for the distinct edge condition.
2017-03-28
Technical Paper
2017-01-1675
Genís Mensa, Núria Parera, Alba Fornells
Abstract Nowadays, the use of high-speed digital cameras to acquire relevant information is a standard for all laboratories and facilities working in passive safety crash testing. The recorded information from the cameras is used to develop and improve the design of vehicles in order to make them safer. Measurements such as velocities, accelerations and distances are computed from high-speed images captured during the tests and represent remarkable data for the post-crash analysis. Therefore, having the exact same position of the cameras is a key factor to be able to compare all the values that are extracted from the images of the tests carried out within a long-term passive safety project. However, since working with several customers involves a large amount of different cars and tests, crash facilities have to readapt for every test mode making it difficult for them to reproduce the correct and precise position of the high-speed cameras throughout the same project.
2017-03-28
Technical Paper
2017-01-1422
Toby Terpstra, Seth Miller, Alireza Hashemian
Abstract Photogrammetry and the accuracy of a photogrammetric solution is reliant on the quality of photographs and the accuracy of pixel location within the photographs. A photograph with lens distortion can create inaccuracies within a photogrammetric solution. Due to the curved nature of a camera’s lens(s), the light coming through the lens and onto the image sensor can have varying degrees of distortion. There are commercially available software titles that rely on a library of known cameras, lenses, and configurations for removing lens distortion. However, to use these software titles the camera manufacturer, model, lens and focal length must be known. This paper presents two methodologies for removing lens distortion when camera and lens specific information is not available. The first methodology uses linear objects within the photograph to determine the amount of lens distortion present. This method will be referred to as the straight-line method.
2017-03-28
Technical Paper
2017-01-1405
Tzu-Sung Wu
Abstract Autonomous Emergency Braking Systems (AEBS) usually contain radar, (stereo) camera and/or LiDAR-based technology to identify potential collision partners ahead of the car, such that to warn the driver or automatically brake to avoid or mitigate a crash. The advantage of camera is less cost: however, is inevitable to face the defects of cameras in AEBS, that is, the image recognition cannot perform good accuracy in the poor or over-exposure light condition. Therefore, the compensation of other sensors is of importance. Motivated by the improvement of false detection, we propose a Pedestrian-and-Vehicle Recognition (PVR) algorithm based on radar to apply to AEBS. The PVR employs the radar cross section (RCS) and standard deviation of width of obstacle to determine whether a threshold value of RCS and standard deviation of width of the pedestrian and vehicle is crossed, and to identity that the objective is a pedestrian or vehicle, respectively.
2017-03-28
Technical Paper
2017-01-1434
Dongran Liu, Marcos Paul Gerardo-Castro, Bruno Costa, Yi Zhang
Abstract Heart rate is one of the most important biological features for health information. Most of the state-of-the-art heart rate monitoring systems rely on contact technologies that require physical contact with the user. In this paper, we discuss a proof-of-concept of a non-contact technology based on a single camera to measure the user’s heart rate in real time. The algorithm estimates the heart rate based on facial color changes. The input is a series of video frames with the automatically detected face of the user. A Gaussian pyramid spatial filter is applied to the inputs to obtain a down-sampled high signal-to-noise ratio images. A temporal Fourier transform is applied to the video to get the signal spectrum. Next, a temporal band-pass filter is applied to the transformed signal in the frequency domain to extract the frequency band of heart beats. We then used the dominant frequency in the Fourier domain to find the heart rate.
2017-03-28
Technical Paper
2017-01-0537
Murat Ates, Ronald D. Matthews, Matthew J. Hall
Abstract A quasi-dimensional model for a direct injection diesel engine was developed based on experiments at Sandia National Laboratory. The Sandia researchers obtained images describing diesel spray evolution, spray mixing, premixed combustion, mixing controlled combustion, soot formation, and NOx formation. Dec [1] combined all of the available images to develop a conceptual diesel combustion model to describe diesel combustion from the start of injection up to the quasi-steady form of the jet. The end of injection behavior was left undescribed in this conceptual model because no clear image was available due to the chaotic behavior of diesel combustion. A conceptual end-of-injection diesel combustion behavior model was developed to capture diesel combustion throughout its life span. The compression, expansion, and gas exchange stages are modeled via zero-dimensional single zone calculations.
2017-03-28
Technical Paper
2017-01-1401
Trong-Duy Nguyen, Joseph Lull, Satish Vaishnav
Abstract In this paper, a method of improving the automated vehicle’s perception using a multi-pose camera system (MPCS) is presented. The proposed MPCS is composed of two identical colored and high frame-rate cameras: one installed in the driver side and the other in the passenger side. Perspective of MPCS varies depending on the width of vehicle type in which MPCS is installed. To increase perspective, we use the maximum width of the host vehicle as camera to camera distance for the MPCS. In addition, angular positions of the two cameras in MPCS are controlled by two separate electric motor-based actuators. Steering wheel angle, which is available from the vehicle Controller Area Network (CAN) messages, is used to supply information to the actuators to synchronize MPCS camera positions with the host vehicle steering wheel.
2017-03-28
Journal Article
2017-01-0927
Carl Justin Kamp, Shawn Zhang, Sujay Bagi, Victor Wong, Greg Monahan, Alexander Sappok, Yujun Wang
Abstract Diesel engine exhaust aftertreatment components, especially the diesel particulate filter (DPF), are subject to various modes of degradation over their lifetimes. One particular adverse effect on the DPF is the significant rise in pressure drop due to the accumulation of engine lubricant-derived ash which coats the inlet channel walls effectively decreasing the permeability of the filter. The decreased permeability due to ash in the DPF can result in increased filter pressure drop and decreased fuel economy. A unique two-step approach, consisting of experimental measurements and direct numerical simulations using ultra-high resolution 3D imaging data, has been utilized in this study to better understand the effects of ash accumulation on engine aftertreatment component functionality.
2017-03-28
Journal Article
2017-01-0674
Benjamin Matthew Wolk, Isaac Ekoto
Abstract Pulsed nanosecond discharges (PND) can achieve ignition in internal combustion engines through enhanced reaction kinetics as a result of elevated electron energies without the associated increases in translational gas temperature that cause electrode erosion. Atomic oxygen (O), including its electronically excited states, is thought to be a key species in promoting low-temperature ignition. In this paper, high-voltage (17-24 kV peak) PND are examined in oxygen/nitrogen/carbon dioxide/water mixtures at engine-relevant densities (up to 9.1 kg/m3) through pressure-rise calorimetry and direct imaging of excited-state O-atom and molecular nitrogen (N2) in an optically accessible spark calorimeter, with the anode/cathode gap distance set to 5 mm or with an anode-only configuration (DC corona). The conversion efficiency of pulse electrical energy into thermal energy was measured for PND with secondary streamer breakdown (SSB) and similar low-temperature plasmas (LTP) without.
2017-03-28
Journal Article
2017-01-0716
Randy Hessel, Zongyu Yue, Rolf Reitz, Mark Musculus, Jacqueline O'Connor
Abstract One way to develop an understanding of soot formation and oxidation processes that occur during direct injection and combustion in an internal combustion engine is to image the natural luminosity from soot over time. Imaging is possible when there is optical access to the combustion chamber. After the images are acquired, the next challenge is to properly interpret the luminous distributions that have been captured on the images. A major focus of this paper is to provide guidance on interpretation of experimental images of soot luminosity by explaining how radiation from soot is predicted to change as it is transmitted through the combustion chamber and to the imaging. The interpretations are only limited by the scope of the models that have been developed for this purpose. The end-goal of imaging radiation from soot is to estimate the amount of soot that is present.
2017-03-28
Journal Article
2017-01-1543
Jonathan Jilesen, Adrian Gaylard, Jose Escobar
Abstract Vehicle rear and side body soiling has been a concern since the earliest cars. Traditionally, soiling has been seen to be less importance than vehicle aerodynamics and acoustics. However, increased reliance on sensors and cameras to assist the driver means that there are more surfaces of the vehicle that must be kept clean. Failure to take this into consideration means risking low customer satisfaction with new features. This is because they are likely to fail under normal operating conditions and require constant cleaning. This paper numerically investigates features known to have an influence on side and rear face soiling with a demonstration vehicle. These changes include rim design, diffuser strakes and diffuser sharpening. While an exhaustive investigation of these features is beyond the scope of this study, examples of each feature will be considered.
2017-03-14
Journal Article
2017-01-9750
Shawn Harrington, Joseph Teitelman, Erica Rummel, Brendan Morse, Peter Chen, Donald Eisentraut, Daniel McDonough
Abstract With the prevalence of satellite imagery in the analysis of collision events growing in the field of accident reconstruction, this research aims to quantify, refine, and compare the accuracies of measurements obtained utilizing conventional instruments to the measurements obtained using Google Earth Pro software. Researchers documented and obtained 1305 unique measurements from 68 locations in 25 states and provinces in the United States, Canada, and Australia using measuring wheels and tape measures. Measurements of relevant features at each location (crosswalks, curved roadways, off-road features, etc.) were documented and subdivided into three groups: On-Road, Off-Road, and Curved Path measurements. These measurements were compared to the measurements obtained of the same features from current and historical satellite imagery within Google Earth Pro.
Viewing 1 to 30 of 366