Junxing ZHANG Shuo YANG Chunjuan BO Huimin LU
Vehicle logo detection technology is one of the research directions in the application of intelligent transportation systems. It is an important extension of detection technology based on license plates and motorcycle types. A vehicle logo is characterized by uniqueness, conspicuousness, and diversity. Therefore, thorough research is important in theory and application. Although there are some related works for object detection, most of them cannot achieve real-time detection for different scenes. Meanwhile, some real-time detection methods of single-stage have performed poorly in the object detection of small sizes. In order to solve the problem that the training samples are scarce, our work in this paper is improved by constructing the data of a vehicle logo (VLD-45-S), multi-stage pre-training, multi-scale prediction, feature fusion between deeper with shallow layer, dimension clustering of the bounding box, and multi-scale detection training. On the basis of keeping speed, this article improves the detection precision of the vehicle logo. The generalization of the detection model and anti-interference capability in real scenes are optimized by data enrichment. Experimental results show that the accuracy and speed of the detection algorithm are improved for the object of small sizes.
Ping LI Feng ZHOU Bo ZHAO Maliang LIU Huaxi GU
This paper presents a large-angle imaging algorithm based on a dynamic scattering model for inverse synthetic aperture radar (ISAR). In this way, more information can be presented in an ISAR image than an ordinary RD image. The proposed model describes the scattering characteristics of ISAR target varying with different observation angles. Based on this model, feature points in each sub-image of the ISAR targets are extracted and matched using the scale-invariant feature transform (SIFT) and random sample consensus (RANSAC) algorithms. Using these feature points, high-precision rotation angles are obtained via joint estimation, which makes it possible to achieve a large angle imaging using the back-projection algorithm. Simulation results verifies the validity of the proposed method.
Futoshi KUROKI Shouta SORA Kousei KUMAHARA
A ring-resonator type of electrode (RRTE) has been proposed to detect the circulating tumor cell (CTC) for evaluation of the current cancer progression and malignancy in clinical applications. Main emphasis is placed on the identification sensitivity for the lossy materials that can be found in biomedical fields. At first, the possibility of the CTC detection was numerically considered to calculate the resonant frequency of the RRTE catching the CTC, and it was evident that the RRTE with the cell has the resonant frequency inherent in the cell featured by its complex permittivity. To confirm the numerical consideration, the BaTiO3 particle, whose size was similar to that of the CTC, was inserted in the RRTE instead of the CTC as a preliminary experiment. Next, the resonant frequencies of the RRTE with internal organs of the beef cattle such as liver, lung, and kidney were measured for evaluation of the lossy materials such as the CTC, and degraded Q curves were observed because the Q-factors inherent in the internal organs were usually low due to the poor loss tangents. To overcome such difficulty, the RRTE, the oscillator circuit consisting of the FET being added, was proposed to improve the identification sensitivity. Comparing the identification sensitivity of the conventional RRTE, it has been improved because the oscillation frequency spectrum inherent in an internal organ could be easily observed thanks to the oscillation condition with negative resistance. Thus, the validity of the proposed technique has been confirmed.
Danlei XING Fei WU Ying SUN Xiao-Yuan JING
Cross-project defect prediction (CPDP) is a feasible solution to build an accurate prediction model without enough historical data. Although existing methods for CPDP that use only labeled data to build the prediction model achieve great results, there are much room left to further improve on prediction performance. In this paper we propose a Semi-Supervised Discriminative Feature Learning (SSDFL) approach for CPDP. SSDFL first transfers knowledge of source and target data into the common space by using a fully-connected neural network to mine potential similarities of source and target data. Next, we reduce the differences of both marginal distributions and conditional distributions between mapped source and target data. We also introduce the discriminative feature learning to make full use of label information, which is that the instances from the same class are close to each other and the instances from different classes are distant from each other. Extensive experiments are conducted on 10 projects from AEEEM and NASA datasets, and the experimental results indicate that our approach obtains better prediction performance than baselines.
Nobuhiko ITOH Takanori IWAI Ryogo KUBO
Road traffic collisions are an extremely serious and often fatal issue. One promising approach to mitigate such collisions is the use of connected car services that share road traffic information obtained from vehicles and cameras over mobile networks. In connected car services, it is important for data chunks to arrive at a destination node within a certain deadline constraint. In this paper, we define a flow from a vehicle (or camera) to the same vehicle (or camera) via an MEC server, as a mission critical (MC) flow, and call a deadline of the MC flow the MC deadline. Our research objective is to achieve a higher arrival ratio within the MC deadline for the MC flow that passes through both the radio uplink and downlink. We previously developed a deadline-aware scheduler with consideration for quality fluctuation (DAS-QF) that considers chunk size and a certain deadline constraint in addition to radio quality and utilizes these to prioritize users such that the deadline constraints are met. However, this DAS-QF does not consider that the congestion levels of evolved NodeB (eNB) differ depending on the eNB location, or that the uplink congestion level differs from the downlink congestion level in the same eNB. Therefore, in the DAS-QF, some data chunks of a MC flow are discarded in the eNB when they exceed either the uplink or downlink deadline in congestion, even if they do not exceed the MC deadline. In this paper, to reduce the eNB packet drop probability due to exceeding either the uplink and downlink deadline, we propose a deadline coordination function (DCF) that adaptively sets each of the uplink and downlink deadlines for the MC flow according to the congestion level of each link. Simulation results show that the DAS-QF with DCF offers higher arrival ratios within the MC deadline compared to DAS-QF on its own
Yasuhiro MOCHIDA Takayuki NAKACHI Takahiro YAMAGUCHI
High frame rate (HFR) video is attracting strong interest since it is considered as a next step toward providing Ultra-High Definition video service. For instance, the Association of Radio Industries and Businesses (ARIB) standard, the latest broadcasting standard in Japan, defines a 120 fps broadcasting format. The standard stipulates temporally scalable coding and hierarchical transmission by MPEG Media Transport (MMT), in which the base layer and the enhancement layer are transmitted over different paths for flexible distribution. We have developed the first ever MMT transmitter/receiver module for 4K/120fps temporally scalable video. The module is equipped with a newly proposed encapsulation method of temporally scalable bitstreams with correct boundaries. It is also designed to be tolerant to severe network constraints, including packet loss, arrival timing offset, and delay jitter. We conducted a hierarchical transmission experiment for 4K/120fps temporally scalable video. The experiment demonstrated that the MMT module was successfully fabricated and capable of dealing with severe network constraints. Consequently, the module has excellent potential as a means to support HFR video distribution in various network situations.
Kenji MII Akihito NAGAHAMA Hirobumi WATANABE
This paper proposes an ultra-low quiescent current low-dropout regulator (LDO) with a flipped voltage follower (FVF)-based load transient enhanced circuit for wireless sensor network (WSN). Some characteristics of an FVF are low output impedance, low voltage operation, and simple circuit configuration [1]. In this paper, we focus on the characteristics of low output impedance and low quiescent current. A load transient enhanced circuit based on an FVF circuit configuration for an LDO was designed in this study. The proposed LDO, including the new circuit, was fabricated in a 0.6 µm CMOS process. The designed LDO achieved an undershoot of 75 mV under experimental conditions of a large load transient of 100 µA to 10 mA and a current slew rate (SR) of 1 µs. The quiescent current consumed by the LDO at no load operation was 204 nA.
Plane wave scattering from a circular conducting cylinder and a circular conducting strip has been formulated by equivalent surface currents which are postulated from the scattering geometrical optics (GO) field. Thus derived radiation far fields are found to be the same as those formulated by a conventional physical optics (PO) approximation for both E and H polarizations.
In this paper, we study a radio frequency (RF)-powered backscatter assisted cognitive radio network (CRN), where an eavesdropper exists. This network includes a primary transmitter, a pair of secondary transmitter and receiver, a friendly jammer and an eavesdropper. We assume that the secondary transmitter works in ambient backscatter (AmBack) mode and the friendly jammer works in harvest-then-transmit (HTT) mode, where the primary transmitter serves as energy source. To enhance the physical layer security of the secondary user, the friendly jammer uses its harvested energy from the primary transmitter to transmit jamming noise to the eavesdropper. Furthermore, for maximizing the secrecy rate of secondary user, the optimal time allocation including the energy harvesting and jamming noise transmission phases is obtained. Simulation results verify the superiority of the proposed scheme.
Reference current used in sense amplifiers is a crucial factor in a single-end read manner for emerging memories. Dummy cell average read scheme uses multiple pairs of dummy cells inside the array to generate an accurate reference current for data sensing. The previous research adopts current mirror sense amplifier (CMSA) which is compatible with the dummy cell average read scheme. However, clamped bit-line sense amplifier (CBLSA) has higher sensing speed and lower power consumption compared with CMSA. Therefore, applying CBLSA to dummy cell average read scheme is expected to enhance the performance. This paper reveals that direct combination of CBLSA and dummy cell average read scheme leads to sense margin degradation. In order to solve this problem, a new array design is proposed to make CBLSA compatible with dummy cell average read scheme. Current mirror structure is employed to prevent CBLSA from being short-circuited directly. The simulation result shows that the minimum sensible tunnel magnetoresistance ratio (TMRR) can be extended from 14.3% down to 1%. The access speed of the proposed sensing scheme is less than 2 ns when TMRR is 70% or larger, which is about twice higher than the previous research. And this circuit design just consumes half of the energy in one read cycle compared with the previous research. In the proposed array architecture, all the dummy cells can be always short-circuited in totally isolated area by low-resistance metal wiring instead of using controlling transistors. This structure is able to contribute to increasing the dummy cell averaging effect. Besides, the array-level simulation validates that the array design is accessible to every data cell. This design is generally applicable to any kinds of resistance-variable emerging memories including STT-MRAM.
Jianmei ZHANG Pengyu WANG Feiyang GONG Hongqing ZHU Ning CHEN
Finding the correspondence between two images of the same object or scene is an active research field in computer vision. This paper develops a rapid and effective Content-based Superpixel Image matching and Stitching (CSIS) scheme, which utilizes the content of superpixel through multi-features fusion technique. Unlike popular keypoint-based matching method, our approach proposes a superpixel internal feature-based scheme to implement image matching. In the beginning, we make use of a novel superpixel generation algorithm based on content-based feature representation, named Content-based Superpixel Segmentation (CSS) algorithm. Superpixels are generated in terms of a new distance metric using color, spatial, and gradient feature information. It is developed to balance the compactness and the boundary adherence of resulted superpixels. Then, we calculate the entropy of each superpixel for separating some superpixels with significant characteristics. Next, for each selected superpixel, its multi-features descriptor is generated by extracting and fusing local features of the selected superpixel itself. Finally, we compare the matching features of candidate superpixels and their own neighborhoods to estimate the correspondence between two images. We evaluated superpixel matching and image stitching on complex and deformable surfaces using our superpixel region descriptors, and the results show that new method is effective in matching accuracy and execution speed.
Faster R-CNN uses a region proposal network which consists of a single scale convolution filter and fully connected networks to localize detected regions. However, using a single scale filter is not enough to detect full regions of characters. In this letter, we propose a simple but effective way, i.e., utilizing variously sized convolution filters, to accurately detect Chinese characters of multiple scales in documents. We experimentally verified that our method improved IoU by 4% and detection rate by 3% than the previous single scale Faster R-CNN method.
Jin LIU Masahide HATANAKA Takao ONOYE
Lately, an increasing number of wireless local area network (WLAN) access points (APs) are deployed to serve an ever increasing number of mobile stations (STAs). Due to the limited frequency spectrum, more and more AP and STA nodes try to access the same channel. Spatial spectrum reuse is promoted by the IEEE 802.11ax task group through dynamic sensitivity control (DSC), which permits cochannel operation when the received signal power at the prospective transmitting node (PTN) is lower than an adjusted carrier sensing threshold (CST). Previously-proposed DSC approaches typically calculate the CST without node grouping by using a margin parameter that remains fixed during operation. Setting the margin has previously been done heuristically. Finding a suitable value has remained an open problem. Therefore, herein, we propose a DSC approach that employs a node grouping method for adaptive calculation of the CST at the PTN with a channel-aware and margin-free formula. Numerical simulations for dense residential WLAN scenario reveal total throughput and Jain's fairness index gains of 8.4% and 7.6%, respectively, vs. no DSC (as in WLANs deployed to present).
Minseok KIM Tatsuki IWATA Shigenobu SASAKI Jun-ichi TAKADA
In radio channel measurements and modeling, directional scanning via highly directive antennas is the most popular method to obtain angular channel characteristics to develop and evaluate advanced wireless systems for high frequency band use. However, it is often insufficient for ray-/cluster-level characterizations because the angular resolution of the measured data is limited by the angular sampling interval over a given scanning angle range and antenna half power beamwidth. This study proposes the sub-grid CLEAN algorithm, a novel technique for high-resolution multipath component (MPC) extraction from the multi-dimensional power image, so called double-directional angular delay power spectrum. This technique can successfully extract the MPCs by using the multi-dimensional power image. Simulation and measurements showed that the proposed technique could extract MPCs for ray-/cluster-level characterizations and channel modeling. Further, applying the proposed method to the data captured at 58.5GHz in an atrium entrance hall environment which is an indoor hotspot access scenario in the fifth generation mobile system, the multipath clusters and corresponding scattering processes were identified.
Yong YANG Junwei LU Baoxian WANG Weigang ZHAO
The concrete quality of supporting layer in ballastless track is important for the safe operation of a high-speed railway (HSR). However, the supporting layer is covered by the upper track slab and the functional layer, and it is difficult to detect concealed defects inside the supporting layer. To solve this problem, a method of elastic wave velocity imaging is proposed to analyze the concrete quality. First, the propagation path of the elastic wave in the supporting layer is analyzed, and a head-wave arrival-time (HWAT) extraction method based on the wavelet spectrum correlation analysis (WSCA) is proposed. Then, a grid model is established to analyze the relationships among the grid wave velocity, travel route, and travel time. A loss function based on the total variation is constructed, and an inverse method is applied to evaluate the elastic wave velocity in the supporting layer. Finally, simulation and field experiments are conducted to verify the suppression of noise signals and the accuracy of an inverse imaging for the elastic wave velocity estimation. The results show that the WSCA analysis could extract the HWAT efficiently, and the inverse imaging method could accurately estimate wave velocity in the supporting layer.
In this letter, we propose a more secure modeling and simulation approach that can systematically detect state variable corruptions caused by buffer overflows in simulation models. Using our approach, developers may not consider secure coding practices related to the corruptions. We have implemented a prototype of the approach based on a modeling and simulation formalism and an open source simulator. Through optimization, the prototype could show better performance, compared to the original simulator, and detect state variable corruptions.
Asuka NAKAJIMA Takuya WATANABE Eitaro SHIOJI Mitsuaki AKIYAMA Maverick WOO
With our ever increasing dependence on computers, many governments around the world have started to investigate strengthening the regulations on vulnerabilities and their lifecycle management. Although many previous works have studied this problem space for mainstream software packages and web applications, relatively few have studied this for consumer IoT devices. As our first step towards filling this void, this paper presents a pilot study on the vulnerability disclosures and patch releases of three prominent consumer IoT vendors in Japan and three in the United States. Our goals include (i) characterizing the trends and risks in the vulnerability lifecycle management of consumer IoT devices using accurate long-term data, and (ii) identifying problems, challenges, and potential approaches for future studies of this problem space. To this end, we collected all published vulnerabilities and patches related to the consumer IoT products by the included vendors between 2006 and 2017; then, we analyzed our dataset from multiple perspectives, such as the severity of the included vulnerabilities and the timing of the included patch releases with respect to the corresponding disclosures and exploits. Our work has uncovered several important findings that may inform future studies. These findings include (i) a stark contrast between how the vulnerabilities in our dataset were disclosed in the two markets, (ii) three alarming practices by the included vendors that may significantly increase the risk of 1-day exploits for customers, and (iii) challenges in data collection including crawling automation and long-term data availability. For each finding, we also provide discussions on its consequences and/or potential migrations or suggestions.
Rengie Mark D. MAILIG Min Gee KIM Shun-ichiro OHMI
In this paper, the effects of the TiN encapsulating layer and the dopant segregation process on the interface properties and the Schottky barrier height reduction of PdEr-silicide/n-Si(100) were investigated. The results show that controlling the initial location of the boron dopants by adding the TiN encapsulating layer lowered the Schottky barrier height (SBH) for hole to 0.20 eV. Furthermore, the density of interface states (Dit) on the order of 1011eV-1cm-2 was obtained indicating that the dopant segregation process with TiN encapsulating layer effectively annihilated the interface states.
Songlin DU Yuan LI Takeshi IKENAGA
High frame rate and ultra-low delay are the most essential requirements for building excellent human-machine-interaction systems. As a state-of-the-art local keypoint detection and feature extraction algorithm, A-KAZE shows high accuracy and robustness. Nonlinear scale space is one of the most important modules in A-KAZE, but it not only has at least one frame delay and but also is not hardware friendly. This paper proposes a hardware oriented nonlinear scale space for high frame rate and ultra-low delay A-KAZE matching system. In the proposed matching system, one part of nonlinear scale space is temporally forward and calculated in the previous frame (proposal #1), so that the processing delay is reduced to be less than 1 ms. To improve the matching accuracy affected by proposal #1, pre-adjustment of nonlinear scale (proposal #2) is proposed. Previous two frames are used to do motion estimation to predict the motion vector between previous frame and current frame. For further improvement of matching accuracy, pixel-level pre-adjustment (proposal #3) is proposed. The pre-adjustment changes from block-level to pixel-level, each pixel is assigned an unique motion vector. Experimental results prove that the proposed matching system shows average matching accuracy higher than 95% which is 5.88% higher than the existing high frame rate and ultra-low delay matching system. As for hardware performance, the proposed matching system processes VGA videos (640×480 pixels/frame) at the speed of 784 frame/second (fps) with a delay of 0.978 ms/frame.
Yi-ze LE Yong FENG Da-jiang LIU Bao-hua QIANG
Metric learning aims to generate similarity-preserved low dimensional feature vectors from input images. Most existing supervised deep metric learning methods usually define a carefully-designed loss function to make a constraint on relative position between samples in projected lower dimensional space. In this paper, we propose a novel architecture called Naive Similarity Discriminator (NSD) to learn the distribution of easy samples and predict their probability of being similar. Our purpose lies on encouraging generator network to generate vectors in fitting positions whose similarity can be distinguished by our discriminator. Adequate comparison experiments was performed to demonstrate the ability of our proposed model on retrieval and clustering tasks, with precision within specific radius, normalized mutual information and F1 score as evaluation metrics.