Search

Counting the number of people is a common and basic computing operation in many applications. Most of the people counting techniques need a sensing device like camera and apply image processing methods to track pedestrians. However, counting people with cameras in private places raises a lot of security and privacy issues. The passive infra-red sensor (PIR) can detect the body temperature of the infrared and thus provides another promising solution. Although a single PIR can easily identify the passing situations (i.e., in or out) of a single person, the signals of a single PIR is not sufficient to identify the complex situations of multiple people. In the paper, we design a people counting device with a PIR array to detect the passing situations and generate data records with higher discriminability. In addition, we apply the machine learning classification methods including the CNN, the RBM+LR, Decision Tree, and NaiveBayes on the collected data records to identify the passing situations. To validate our design, we conduct experiments to study the feasibility and classification performance and explore the impact factors. The experimental results show that the CNN outperforms the other and achieves the best accuracy, i.e., about 92%. Also, the results show that the captured data records of the PIR array contain sufficient characteristics for identifying complex passing situations and the configuration of the PIR array including the sensor direction and the field of view (FOV) of a PIR modified by the metal tape can significantly impact the discriminability of the collected data.

It is hard to detect on-road objects under various lighting conditions. To improve the quality of the classifier, three techniques are used. We define subclasses to separate daytime and nighttime samples. Then we skip similar samples in the training set to prevent overfitting. With the help of the outside training samples, the detection accuracy is also improved. To detect objects in an edge device, Nvidia Jetson TX2 platform, we exert the lightweight model ResNet-18 FPN as the backbone feature extractor. The FPN (Feature Pyramid Network) generates good features for detecting objects over various scales. With Cascade R-CNN technique, the bounding boxes are iteratively refined for better results., Comment: Accepted at IEEE 21st International Workshop on Multimedia Signal Processing (MMSP 2019)

Multi-Processor Systems-on-Chips (MPSoCs) with 3D-stakced memories frequently work under thermal emergent status due to its high power density. Several thermal-aware task allocation or data placement methods have been proposed for 3D ICs to reduce the number of time-consuming dynamic thermal managements techniques being invoked. However, we observe that, these thermal-aware software designs all consider task allocation or data placement only. Studies show that, with the increasing number of stacked memories and the widening of vertical buses, heat generated by memories is comparable to processors, and synergistically performing thermal control on both processors and memories is a must since vertically aligned modules have the greatest thermal impacts to each other. So, in this paper, we propose the first thermal-aware task and data co-allocation method for MPSoCs with 3D-stacked memories. The proposed method synergistically places data and task considering the heterogeneity of cores and memories to optimize system performance under the given thermal constraint. Among all our test cases, compared to a performance-aware software design, the proposed method has at most 26% performance degradation while the system temperature are kept under the threshold and the performance-aware method has 108.4°C over the threshold. Compared to thermal-aware design that respectively considers data allocation and task allocation only, the proposed method achieves 9.76% of performance improvements on the average.

Enhancement-mode In0.53Ga0.47As n-type metal-oxide-semiconductor field-effect transistors (MOSFETs), with barium zirconate titanate (BZT) and titanium dioxide (TiO2) high-κ materials prepared via the solution–gelation process as gate dielectrics, have been fabricated. The dielectric constants of BZT and TiO2 are 6.67 and 19.3, respectively. The In0.53Ga0.47As MOSFET with TiO2 exhibits better electrical characteristics than the In0.53Ga0.47As MOSFET with BZT. These characteristics include higher maximum drain current density, higher maximum transconductance, and smaller subthreshold swing.

Traffic behavioral monitoring within urban intersections is an essential issue in the Intelligent Transportation Systems (ITS) for a smart city. This paper investigates on gathering traffic information within an urban intersection where accidents frequently occur. In this paper, traffic pattern modeling, trajectory classification and a real-time vehicle tracker within the urban intersection are proposed. Trajectories of vehicles within the intersection are more regular than that of pedestrians; such monotonous trajectories can be classified with Hidden Markov Model (HMM) into various kinds of motion patterns and tracklet prediction can be performed then. So, given an identified prefix trajectory (for a new coming vehicle), the most likely model is determined and the probable template (tracklet) with the highest similarity is selected. This template gives the direction to forecast the next few locations the vehicle may pass through. Besides, tracking all of trajectories in real-time is a computational challenge, on the basis of vehicle movement and tracklet prediction, the proposed method can remove most of the unnecessary particles. The experimental results demonstrate both the computational effectiveness and tracking correctness of the proposed method, and the tracker truly executes in real-time for the intersections of six traffic lanes, say around six vehicles per second on tracking.

Multi-layered micro/nanostructures enable the fine-tuning of spectral bandwidth induced by coherent interference of multi-scattering and refraction bounded by a motive of Fabry Perot etalon. In this paper a ZnO micro/nanostructured thin film device is grown on a 21 nm silver film-sputtered sapphire substrates (0001) and attempted to be complied as a transmission-mode white-light spectrum tuning device. In details, the growth of the structures as a function of temporal duration and the effects of deposition of a 21 nm silver film atop sapphire substrates to the structural and optical properties of ZnO structures are investigated. Results of SEM imaging reveals that the insertion of the silver film enables synthesis of round ZnO particles at micro or nanoscales depending on the duration of growth time as opposed to the floral pattern associated with plain sapphire substrates. Characterizing material crystallization by XRD spectra found one dominant and two minor peaks along the planes of crystallization (002), and (100) and (101), correspondingly. Also, the characteristic PL peak at 380 nm was confirmed when excited at 325 nm, and the intensity decreases as the duration of growth time increases. Finally, growth duration-dependent modulation of peak wavelength and spectral bandwidth of the continuum laser was demonstrated.

Research on polymer solar cells (PSC) using organic p-type (donor) and n-type (acceptor) semiconductors has attracted tremendous scientific and industrial interest in recent years. The charge generation and charge transport play equally important roles in determining the device efficiency. To dramatically increase the area of the donor–acceptor interface for efficient charge separation, a bulk heterojunction (BHJ) is adopted to form an interpenetrating network of donor and acceptor materials. This configuration decreases the distance that excitons need to travel to reach the heterojunction interface, thus reducing exciton recombination. However, the donor and acceptor are randomly interspersed; pathways for charges to reach the electrodes through the active layer are disordered. Free charges are likely to encounter an opposite charge, resulting in charge recombination and reduced current. 4] Moreover, space charge may be built up if charges are locally trapped on isolated domains. Furthermore, an increase in the thickness of the BHJ layer to enhance absorption is usually accompanied by deteriorated charge collection. Consequently, controlling phase separation toward optimal morphology in BHJ by external treatments, such as thermal or solvent annealing, is an important but challenging task. To provide a direct path for charge transport while maintaining a large interfacial area, the ideal architecture of the donor and acceptor is the periodic, vertically aligned, and interpenetrating ordered bulk heterojunction (OBHJ). The electrons and holes have straight and independent pathways to the electrodes to shorten the carrier transport length and reduce the probability of charge recombination. Several elegant studies have attempted to demonstrate this conceptual architecture, for example by a template-assisted strategy or self-assembly of block copolymer. However, realization of high-performance OBHJ devices has not been successful. We envision that designing a system that combines a BHJ for efficient charge generation with an OBHJ for efficient charge transport and collection would be a more practical strategy. Such a configuration is specifically suitable for solar cells with inverted architecture, because an electron-selective layer is required at the bottom of the active layer for electron extraction and hole blocking. For instance, the upper BHJ active layer of an inverted solar cell is infiltrated into vertically aligned nanorods extending from a bottom layer of an inorganic semiconductor (e.g. ZnO or TiO2). [18–20] However, owing to the poor electrical coherence at the organic/inorganic interface, the improvement in efficiency is moderate (PCE ranges from 2.1 to 2.7%). Recently, we reported a cross-linkable fullerene material, [6,6]-phenyl-C61-butyric styryl dendron ester (PCBSD). The formation of a cross-linked PCBSD (C-PCBSD, Figure 1a) planar layer allows realization of a multilayer inverted device by all-solution processing. By using indene–C60 bisadduct (ICBA, Figure 1a) with a higher-lying lowest unoccupied molecular orbital (LUMO) energy level to serve as the acceptor in the blend, an inverted solar cell device based on the ITO/ZnO/C-PCBSD/ICBA:P3HT/PEDOT:PSS/Ag configuration achieved an enhanced power conversion efficiency

Location management is an important issue for object tracking applications of wireless sensor networks. Since locations obtained by most positioning techniques are inherently imprecise, users may send approximate location queries with precision constraints to trade for energy consumption of sensor nodes. Therefore, we propose an Adaptive Location Management scheme called ALM to process approximate location queries. ALM employs a two-tier storage architecture (i.e., centric storage node and local storage node) to facilitate approximate query resolving and to reduce the energy consumption of sensor nodes. We propose a storage node relocation and replication mechanism, which can create, remove replicas of storage nodes and adjust the positions of storage nodes according to the location queries and updates. As such, the number of forwarding location update and query messages is reduced by the proposed storage node relocation and replication mechanism, thereby conserving more energy. The experimental results show that compared with EASE (Xu et al. (2008) [18]), ALM is able to reduce the number of transmission messages and the energy consumption of sensor network, and prolong the network lifetime.

Background: Hand eczema and other skin diseases have been associated with the exposure of chromium among cement workers. Studies on skin disease and other factors associated with the body burden of chromium are limited. Objective: The present study investigated the role of skin disease and smoking in the association with body burden of chromium among cement workers. Methods: Forty five workers (38 men and 7 women) were recruited for this study and interviewed to obtain information on demographic status, lifestyle, employment history, and affecting factors. Urine samples were collected to measure the urinary chromium concentration to represent the body burden of chromium. Results: The average urinary chromium concentration was approximately 6 times higher in non-smoking workers with hand eczema than in non-smoking workers with no skin disease (45.5 vs. 7.6 μg/L). The average chromium level increase to 87.0 μg/L for smokers with the disease. Compared with workers with no hand eczema, the odds ratio of having urinary chromium concentration exceeding the biological exposure index level significantly increased to 11.6 (95% CI = 1.3–102.2) for non-smoking workers with skin disease, and to 48.0 (95% CI = 4.5–510.8) for smoking workers with skin disease. The multiple regression analysis showed that the use of gloves may reduce significantly the chromium exposure. Conclusions: Inadequate protection and personal behavior increase the internal dose of chromium in cement workers. Total body burden of chromium are higher among cement workers with skin disease and smoking habit. These workers deserve intervention education on personal hygiene to reduce the exposure of chromium.

The IEEE 802.11e MAC standard can be used to support the quality of service (QoS) for wireless networks, but the bandwidth resource management is not efficient enough for the wireless multimedia network application. In this paper, a novel cross-layer architecture based on both the bandwidth measurement and queue adaptation is proposed to improve and adjust the video transmission over wireless multimedia networks. In other words, an average bandwidth measurement approach is used to adjust the video delivery immediately. Furthermore, the proposed cross-layer approach uses a dynamic queue adaptation for the current communication protocol of wireless multimedia networks. In addition, it is shown by NS2 simulations that the method of average bandwidth measurement does not consume any network resource, and the proposed method with dynamic queue adaptation can provide a more acceptable service quality by dynamically adapting the transmission rate.

The optical proximity correction (OPC) model and post-OPC verification that takes the developed photoresist (PR) 3D profile into account is needed in the advanced 2Xnm node. The etch process hotspots caused by poor resist profile may not be fully identified during the lithography inspection but will only be observed after the subsequent etch process. A complete mask correction that targets to final etch CD requires not only a lithography R3D profile model but also a etch process compact model. The drawback of existing etch model is to treat the etch CD bias as a function of visibility and pattern density which do not contain the information of resist profile. One important factor to affect the etch CD is the PR lateral erosion during the etch process due to non-vertical PR side wall angle (SWA) and anisotropy of etch plasma source. A simple example is in transferring patterns from PR layer to thin hard mask (HM) layer, which is frequently used in the double pattern (DPT) process. The PR lateral erosion contributes an extra HM etch CD bias which is deviated from PR CD defined by lithography process. This CD bias is found to have a nontrivial dependency on the PR profile and cannot be described by the pattern density or visibility. In this report, we study the etch CD variation to resist SWA under various etch conditions. Physical effects during etch process such as plasma ion reflection and source anisotropy, which modify the local etch rate, are taken into considerations in simulation. The virtual data are generated by Synopsys TCAD tool Sentaurus Topography 3D using Monte Carlo engine. A simple geometry compact model is applied first to explain the behavior of virtual data, however, it works to some extent but lacks accuracy when plasma ion reflection comes into play. A modified version is proposed, for the first time, by including the effects of plasma ion reflection and source anisotropy. The new compact model fits the nonlinear etch CD bias very well for a wide range of resist SWAs from 65 to 90 degrees, which covers the resist profile diversities in most real situations. This result offers a potential application for both resist profile aware and etch process aware mask correction model in the mask synthesis flow.

Pedestrian and motorbike detection are two important areas in obstacle detection on road. Most state-of-the-art detectors are constructed with new features or learning methods on Histograms of Oriented Gradients (HOG) features. However, few researches focus on analyzing which features are complementary for the aforementioned detection. According to our study of pedestrians and motorbikes, there are three major properties including shape, texture, and self-similarity. We design a Shape, Texture and Self-Similarity (STSS) feature for these properties. The features we have employed here are HOG, Local Oriented Pattern (LOP), Color Self-Similarity (CSS), and Texture Self-Similarity (TSS). The STSS detector which combines Shape, Texture, and Self-Similarty features achieves 31% log-average miss rate. At the same time, 93% detection rate at 10-4 false positives per window on INRIA Person Dataset has also been concluded. Besides, we also have evaluated our detector on Caltech Motorbike Dataset and Caltech Pedestrian Dataset, and found the detector outperforms HOG detector in these datasets. As a result, we have shown that these features are complement to each other and useful in pedestrian and motorbike detection.

Pedestrian detection is an important part of intelligent transportation systems. In the literature, Histogram of Oriented Gradients (HOG) detector for pedestrian detection is known for its good performance, but there are still some false detections appearing in the cases with flat area or clustered background. To deal with these problems, in this research work we develop a new feature which is based on pairing comparison computations, called Comparison of Granules (CoG). The idea of CoG is to encode the textural information of local area describing how different the pixel intensities are distributed within a region. It is shown that the special characteristics of CoG feature are “small” and “efficiency” relative to HOG. By incorporating this new feature, we propose a HOG-CoG detector which through our validation experiment achieves 38% log-average miss rate in full image evaluation and 90% detection rate at 10−4 false positives per window on INRIA Person Dataset. Another contribution of this work is that, we also present a training scheme that can be applied on huge database for training a detector. Such training scheme can reduce the number of hard samples during bootstrap training.