Search

It is a big challenge to construct large-scale, high-resolution and high-performance inkjet-printed metal oxide thin film transistor (TFT) arrays with independent gates for the new printed displays. Here, a self-confined inkjet printing technology has been developed to construct large-area (64 × 64 array), high-resolution and high-performance metal oxide bilayer (In2O3/IGZO) heterojunction TFTs with independent bottom gates on transparent glass substrates. Inkjet printing In2O3 dot arrays with the diameters from 55 to 70 μm and the thickness of ∼10 nm were firstly deposited on UV/ozone treated AlOx dielectric layers, and then IGZO dots were selectively printed on the top of In2O3 dots by self-confined technology to form In2O3/IGZO heterojunction channels. When the inkjet-printed IO layers treated by UV/ozone for more than 30 min or oxygen plasma for 5 min prior to print IGZO thin films, the mobility of the resulting printed In2O3/IGZO heterojunction TFTs are correspondingly enhanced to be 18.80 and 28.44 cm2 V−1 s−1 with excellent on/off ratios (>108) and negligible hysteresis. Furthermore, the printed N-Metal-Oxide-Semiconductor (NMOS) inverter consisted of an In2O3/IGZO TFT and an IGZO TFT has been demonstrated, which show excellent performance with the voltage gain up to 112. The strategy demonstrated here can be considered as general approaches to realize a new generation of high-performance printed logic gates, circuits and display driving circuits.

There is rich knowledge in expert systems that can be used to solve practical problems, but its promotion and application must rely on information facilities. The application of both computers and the Internet for Chinese farmers are not common, which leads to restrictions on the promotion and application of expert systems in rural areas of China. On the other hand, the existing call centers lack a professional knowledge base and the method of automatically calling the knowledge base in real-time, which makes it difficult to meet the needs of users wanting to obtain knowledge in a timely manner. To address these problems, a call center embedded in an expert system inference algorithm and knowledge base for farmers to obtain agricultural knowledge through mobile phones or fixed-line telephones was established. By studying the event-condition-action-based (ECA-based) database triggering model, remote method invocation-based (RMI-based) communication and iterative dichotomiser 3 algorithm-based (ID3-based) parameter extraction, the cohesion between the call center and the expert system was realized. The agricultural knowledge audio acquisition model was then coupled with the call center and the expert system was constructed, allowing farmers to acquire agricultural knowledge through mobile phones or fixed phones with fast responses. When used for cotton disease diagnosis, it can achieve a high diagnostic success rate (above 75%) when at least three disease symptoms are input into the expert system via the voice call, which provides an effective channel for Chinese farmers to obtain agricultural knowledge. It presents good application prospects in China, where 5G technology is currently developing rapidly.

In this study, a cotton disease diagnosis method that uses a combined algorithm of case-based reasoning (CBR) and fuzzy logic was designed and implemented. It focuses on the prevention, diagnosis and control of diseases affecting cotton production in China. Conventional methods of disease diagnosis are primarily based on CBR with reference to user-provided symptoms; however, in most cases, user-provided symptoms do not fully meet the requirements of CBR. To address this problem, fuzzy logic is incorporated into CBR to allow for more flexible and accurate models. With the help of CBR and fuzzy reasoning, three diagnostic results can be obtained by the cotton disease diagnosis system (CDDS) constructed in this study: success, success but not exact and failure. To verify the reliability of the CDDS and its ability to diagnose cotton diseases, its diagnostic accuracy and stability were analyzed and compared with the results obtained by the traditional expert scoring method. The analysis results reveal that the CDDS can achieve a high diagnostic success rate (above 90%) and better diagnostic stability than the traditional expert scoring method when at least four disease symptoms are input. The CDDS provides an independent and objective source of information to assist farmers in the diagnosis and prevention of cotton diseases.

Leafy vegetables are consumed in most daily xdiets worldwide. As living standards improve, food quality, safety requirements, and nutrition are becoming increasingly important to consumers when purchasing leafy vegetables. This study proposes an evaluation and traceability method that can be used to track the nutritional quality of leafy vegetables. Employing the principles of the Hazard Analysis and Critical Control Point (HACCP) system combined with fault tree analysis (FTA), a traceability model for the entire production and sale process of leafy vegetables is constructed. Four common leafy vegetables, spinach, rape, lettuce, and celery are examined in this research to establish a nutritional quality index system using fuzzy mathematics subordinate function method to evaluate nutritional quality. A nutritional quality and safety traceability system based on browser/server architecture and quick response (QR) code is then designed and developed for full traceability of leafy vegetable quality. This method can ensure food safety and hygiene through the control of key factors affecting food safety throughout the entire supply chain process.

There is a global shortage of fresh water to meet the clean production of field crops. Water-saving irrigation can increase the production of field crops, but under certain conditions it can be overly water-saving and unsustainable, particularly in complex terrains in China where there is already a shortage of water resources. In light of the insufficient water resources and partial inequality in China, several investigators in China and abroad have researched irrigation nozzles and pumps, but there is little information on the impact of the environment on crop irrigation. The purpose of this study was to provide a fast and effective crop water demand prediction model to estimate the water requirement during the growth period of crops, to achieve the purpose of applying water according to the demand of the crops, and to optimize the irrigation network to determine the optimal network structure in order to reduce irrigation energy consumption in the process. Specifically, this study analyzed the environmental factors, namely, the soil moisture, soil electrical conductivity, air temperature, and light intensity, affecting the physiological water demand of field crops. Relevant sensors were used in the field, and environmental information was collected. This study used the collected environmental data to build a water demand prediction model based on the back propagation (BP) neural network. It also carried out optimized layout and hydrodynamic analyses of the drip irrigation network and identified the best pipe network arrangement. On the basis of the BP neural network, the remarkable R amount of 0.98963 was obtained for the water demand prediction model. The mean square error (MSE) value was 0.00857724. We also found that the drip irrigation pipe network in the H-shaped arrangement was more suitable for field crop irrigation than the comb-shaped and fish bone-shaped arrangements. This study employed the common field crops in China as the research object, used the wireless sensing technology to obtain the field environment information quickly, and constructed a water demand prediction model with high prediction accuracy. Finally, this study realized the precise control of field crop irrigation, reduced the waste of agricultural water, and achieved sustainable field crop production.

Against the background of the depletion of fossil fuel energy sources and environmental degradation, solar greenhouses represent an efficient and sustainable agricultural production method in China. The precise measurement and control of the position of the rolling shutter and rolling film is not only one of the core issues in the modification of the greenhouse environment in actual production but also the basis for improving the cleaner production efficiency of solar greenhouses. In this study, a mathematical algorithm for the control of a rolling shutter and rolling film was developed and is based on the simplification of the circular arc of the greenhouse surface. The “Liaoshen-IV” solar greenhouse was used as an example to illustrate the calculation steps of the algorithm. Any position on the light surface can be determined by calculating the angle at the bottom of the back wall of the greenhouse so that the position of the rolling shutter and the rolling film can be measured accurately by the angle sensor. Experiments using a scale model with one or two surface segments were conducted to demonstrate the advantages of the control algorithm with regard to the accuracy and stability compared with the existing control method of a time-delay relay. The relative errors of the control algorithm were stable and less than 1%. The use of the algorithm improved the control accuracy of the rolling shutter and rolling film and the average relative error was reduced by 6.37% in the first stage and by 3.40% in the second stage compared with the time-delay relay. The precise measurement and control of the position of the rolling shutter and rolling film in a solar greenhouse demonstrated in this study has great significance for practical applications. The proposed method may result in a reduction in energy consumption and greenhouse gas (GHG) emissions in greenhouse environments, thereby contributing to the improvement of resource efficiency and a cleaner production. The study also provides application examples and theoretical references for the automation of rolling shutters and rolling film in greenhouses worldwide.

Multi-player pursuit–evasion games are crucial for addressing the maneuver decision problem arising in the cooperative control of multi-agent systems. This work addresses a particular pursuit–evasion game with three players, Target, Attacker, and Defender. The Attacker aims to capture the Target, while avoiding being captured by the Defender and the Defender tries to defend the Target from being captured by the Attacker, while trying to capture the Attacker at an opportune moment. A two-pronged pursuit–evasion problem in this game is considered and we focus on two aspects: the cooperation between the Target and Defender and balancing the roles of the Attacker between pursuer and evader. A barrier based on the explicit policy method and geometric analysis method is constructed to separate the whole state space into two disjoint parts that correspond to two winning regions for the Attacker and Target–Defender team. The main contributions of this work are obtaining the players’ winning regions and providing a complete game solution by analyzing the optimal strategies and trajectories of the players based on the barrier.

A portable and flexible self-powered biosensor based on ZnO nanowire arrays (ZnO NWs) and flexible PET substrate has been designed and fabricated for real-time monitoring in swimming. Based on the piezoelectric effect of polar ZnO NWs, the fabricated biosensor can work in both air and water without any external power supply. In addition, the biosensor can be easily attached to the surface of the skin to precisely monitor the motion state such as joint moving angle and frequency during swimming. The constant output piezoelectric signal in different relative humidity levels enables actual application in different sports, including swimming. Therefore, the biosensor can be utilized to monitor swimming strokes by attaching it on the surface of the skin. Finally, a wireless transmitting application is demonstrated by implanting the biosensor in vivo to detect angiogenesis. This portable and flexible self-powered biosensor system exhibits broad application prospects in sport monitoring, human–computer interaction and wireless sport big data.

A model-based offline policy iteration (PI) algorithm and a model-free online Q-learning algorithm are proposed for solving fully cooperative linear quadratic dynamic games. The PI-based adaptive Q-learning methodcan learn the feedback Nash equilibrium online using the state samples generated by behavior policies, without sending inquiries to the system model. Unlike the existing Q-learning methods, this novel Q-learning algorithm executes both policy evaluation and policy improvement in an adaptive manner. We prove the convergence of the offline PI algorithm by proving its equivalence to Newtons method while solving the game algebraic Riccati equation (GARE). Furthermore, we prove that the proposed Q-learning method will converge to the Nash equilibrium under a small learning rate if the method satisfies certain persistence of excitation conditions, which can be easily met by suitable behavior policies. Our simulation results demonstrate the good performance of the proposed online adaptive Q-learning algorithm.

Improving the extraction of cucumber leaf spot disease under complex backgrounds.Redefining the feature distance between pixel xj and clustering center vi.Calculating the two-dimensional neighborhood mean gray value as a sample point.Proposing a new weighting method for gray value and neighborhood gray value. Research reported in this paper aims to improve the extraction of cucumber leaf spot disease under complex backgrounds. An improved fuzzy C-means (FCM) algorithm is proposed in this paper. First, three runs of the marked-watershed algorithm, based on HSI space, are applied to isolate the target leaf. Second, the distance between the pixel xj and the cluster center vi is defined as xj2-vi2. Third, the pixel's neighborhood mean gray value, which constitutes a two-dimensional vector with grayscale information, is calculated as a sample point, rather than FCM grayscale. Finally, the neighborhood mean gray value and pixel gray value are weighted by matrix w. To evaluate the robustness and accuracy of the proposed segmentation method, tests were conducted for 129 cucumber disease images in vegetable disease database. Results show that average segmentation error was only 0.12%. The proposed method provides an effective and robust segmentation means for sorting and grading apples in cucumber disease diagnosis, and it can be easily adapted for other imaging-based agricultural applications.

China is the world's largest energy consumer and is also leading in the solar photovoltaic industry. The solar energy industry, with advantageous clean and highly efficient production, is integral in maintaining the sustainable energy supply in an environmentally sensitive manner. Further development of the Chinese photovoltaic industry requires a status assessment for the current industrial setting. An analysis on the 58 Chinese photovoltaic listed enterprises is conducted in this study to analyze the operating performance, industry agglomeration and spatial characteristics of Chinese photovoltaic industry. Comprehensive analysis and evaluation of the enterprises’ operating performance are based on financial data by utilizing the Data Envelopment Analysis (DEA) method; Systematic analysis examines spatial characteristics through the application of the spatial autocorrelation analysis method. Results confirm the Chinese photovoltaic industry preliminarily appeared industry agglomeration with enterprises mainly distributed in east and north China, and gathered in developed coastal provinces. Superior performing photovoltaic enterprises are specifically located in north China. Most of the enterprises are in the up-stream or mid-stream of the photovoltaic industrial chain. Overall operating performance of the photovoltaic industry in China suffers from weak profits due to low technical efficiency. The key to improve the future development of these enterprises is the improvement of operating performance by strengthening technical efficiencies. Concurrently, governmental guidance should address profitable photovoltaic industry investments and improvement of production rates.

In this paper, a policy iteration-based Q-learning algorithm is proposed to solve infinite horizon linear nonzero-sum quadratic differential games with completely unknown dynamics. The Q-learning algorithm, which employs off-policy reinforcement learning (RL), can learn the Nash equilibrium and the corresponding value functions online, using the data sets generated by behavior policies. First, we prove equivalence between the proposed off-policy Q-learning algorithm and an offline PI algorithm by selecting specific initially admissible polices that can be learned online. Then, the convergence of the off-policy Q-learning algorithm is proved under a mild rank condition that can be easily met by injecting appropriate probing noises into behavior policies. The generated data sets can be repeatedly used during the learning process, which is computationally effective. The simulation results demonstrate the effectiveness of the proposed Q-learning algorithm.

Plasmon in two-dimensional electron gas (2DEG) has long been considered as a promising active medium for terahertz emitters and detectors. However, the efficiency of terahertz plasmonic devices is severely limited by the high damping rate of plasma wave in solid state. In addition to the enhancement of plasmon lifetime by using 2DEGs with higher carrier mobility, engineering on the boundary condition and electromagnetic environment of plasmon cavity helps to preserve the plasmon states. Here we report on terahertz reflection spectroscopy of plasmon states in a grating-coupled AlGaN/GaN-2DEG plasmonic device at 7 K in equilibrium with ambient blackbody irradiation. Localized plasmon states and plasmon-polariton states were observed when the core plasmonic device is integrated with a silicon lens and when it is embedded in a terahertz Fabry-Perot cavity, respectively. Simulation results including the reflection spectra and total reflection power agree well with the measured results. The Rabi splitting is found to be inversely proportional to the resonance frequency, and follows a linear relation with the square root of the sheet electron density. A normalized coupling ratio, ΩRω0≈0.13, is achieved between the Rabi splitting ΩR and the resonance frequency ω0. The coupling ratio could be further increased to allow for ultrastrong coupling between terahertz photons and plasmons.

Gastric cancer (GC) is a multifactorial process, accompanied by alterations in metabolic pathways. Non-invasive metabolic profiling facilitates GC diagnosis at early stage leading to an improved prognostic outcome. Herein, mesoporous PdPtAu alloys are designed to characterize the metabolic profiles in human blood. The elemental composition is optimized with heterogeneous surface plasmonic resonance, offering preferred charge transfer for photoinduced desorption/ionization and enhanced photothermal conversion for thermally driven desorption. The surface structure of PdPtAu is further tuned with controlled mesopores, accommodating metabolites only, rather than large interfering compounds. Consequently, the optimized PdPtAu alloy yields direct metabolic fingerprints by laser desorption/ionization mass spectrometry in seconds, consuming 500 nL of native plasma. A distinct metabolic phenotype is revealed for early GC by sparse learning, resulting in precise GC diagnosis with an area under the curve of 0.942. It is envisioned that the plasmonic alloy will open up a new era of minimally invasive blood analysis to improve the surveillance of cancer patients in the clinical setting.

Freshwater aquaculture is the main form of aquaculture in inland areas, and pollution is a significant challenge. The aim of this work was to propose an effective risk assessment method for aquaculture to predict pollutant bioaccumulation in fish tissue based on water environment variables. This method was designed to assess the risk of pollutant levels in fish tissue that exceed China’s food safety standards. Firstly, field observation and investigation were conducted to analyze the migration mechanisms of the main pollutants between fish tissue and the water environment. Secondly, a classification and regression tree (CART) and generalized additive model (GAM) nexus method was used to analyze the environmental variables related to pollutants in the water environment and establish prediction models. Finally, models of the bio-reaction kinetics were established to predict the pollutant bioaccumulation in fish tissue. The experimental results indicate that the GAM models could explain 73.7%, 84.4%, and 84.5% of the deviations for the three predicted pollutants. For 90% of the sample data, the accuracy of the bioaccumulation prediction was 93.44% for volatile phenol bioaccumulation, and 90.16% for Cu and coliform bioaccumulation. The proposed risk assessment method could be extended for aquaculture applications that require monitoring of additional parameters to safeguard aquaculture quality and facilitate effective safety risk management.

This paper addresses a particular pursuit-evasion game with two pursuers with slower speed but smaller minimum turning radius and a faster evader with bigger minimum turning radius. This is a game of kind, and what interests us is how to construct the barrier that separates the state zone into capture zone and escape zone, and what the optimal strategies for the players are on the barrier. Under some mild assumptions, we give the explicit form of the barrier near the BUP (i.e., the boundary of the usable part on the boundary of the target set) by using Isaacs’ method, and a procedure to construct the barrier when the retrogressive time is big enough by determining the optimal strategies for the players on the BUP. Then we prove that the optimal strategies remain unchanged near some special parts on the BUP, and we give two examples to illustrate these situations.

We report an attempt to improve the sensitivity of terahertz detection based on self-mixing in antenna-coupled field-effect transistors by enhancing the field-effect factor and the antenna factor with a reduced gate length and a reduced antenna-gate gap, respectively. An optical noise equivalent power (NEP) of 3.7 pW/Hz at 0.65 THz was achieved in a GaN/AlGaN high-electron-mobility transistor (HEMT) with a gate length of 300 nm and an antenna-gate gap of 200 nm at room temperature. It was found that the antenna factor was inversely proportional to the antenna-gate gap, and the responses upon coherent/incoherent terahertz irradiation were well described by the self-mixing model. To fill the NEP gap of 0.1−1 pW/Hz between room-temperature and cryogenic detectors by HEMT-based detectors at room temperature, impedance match needs to be carefully considered.

Aiming at the requirement of passive terahertz imaging, we report a high-sensitivity terahertz detector based on an antenna-coupled AlGaN/GaN high-electron-mobility transistor (HEMT) at 77 K without using low-noise terahertz amplifier. The measured optical noise-equivalent power and the noise-equivalent temperature difference of the detector were about 0.3 p W / H z and 370 mK in a 200 ms integration time over a bandwidth of 0.7 − 0.9 THz, respectively. By using this detector, we demonstrated passive terahertz imaging of room-temperature objects with signal-to-noise ratio up to 13 dB. Further improvement in the sensitivity may allow passive terahertz imaging using AlGaN/GaN-HEMT at room temperature.

In the process of agricultural production, the yield of field crops is affected by the interaction of various factors. This relationship between soil nutrients, fertilization, and yield is very complicated. It has strongly non-linear and black box characteristics and is difficult to quantify with traditional analytic methods. This paper proposes a method for the precision fertilization of maize based on the wavelet-BP neural network. Firstly, the existing “3414″ experimental data was chosen and interpolated as the performing data for the modeling. Then, the general condition of low frequency and details of high-frequency of the yield were calculated via the wavelet decomposition and reconstruction method. After that, the wavelet-BP neural network was established. Three different BP neural sub-networks for each component of the yield after wavelet analysis were established, and the output of each sub-network was summed to obtain the predicted corn yield. The results show that the model that combines wavelet analysis with the BP neural network achieves better performance than the traditional BP neural network, Support Vector Machine (SVM), and Random Forest regarding accuracy and stability, which indicates the feasibility of this method. Finally, the best fertilizer amount that achieves the maximum yield or maximum profit is calculated based on nonlinear programming. The proposed method of the wavelet-BP neural network provides a new technique for precision fertilization research and enriches the existing fertilization system. The application of precision fertilization based on the wavelet-BP neural network has important practical significance that guides and increases maize production, while reducing production cost and agricultural pollution. At the same time, it also provides a way to solve similar problems in the entire field of cleaner agricultural production.

In this paper, a novel online Q-Iearning approach is proposed to solve the Infinite Horizon Linear Quadratic Regulator (IHLQR) problem for continuous-time (CT) linear time-invariant (LMI) systems. The proposed Q-Iearning algorithm employing off-policy reinforcement learning (RL) technology improves the exploration ability of Q-Iearning to the state space. During the learning process, the Q-Iearning algorithm can be implemented just using the data sets which just contains the information of the behavior policy and the corresponding system state, thus is data- driven. Moreover, the data sets can be used repeatedly, which is computationally efficient. A mild condition on probing noise is established to ensure the converge of the proposed Q-Iearning algorithm. Simulation results demonstrate the effectiveness of the developed algorithm.

Indocyanine green (ICG) is a near-infrared dye that has been used in the clinic for retinal angiography, and defining cardiovascular and liver function for over 50 years. Recently, there has been an increasing interest in the incorporation of ICG into nanoparticles (NPs) for cancer theranostic applications. Various types of ICG-incorporated NPs have been developed and strategically functionalised to embrace multiple imaging and therapeutic techniques for cancer diagnosis and treatment. This review systematically summaries the biodistribution of various types of ICG-incorporated NPs for the first time, and discusses the principles, opportunities, limitations, and application of ICG-incorporated NPs for cancer theranostics. We believe that ICG-incorporated NPs would be a promising multifunctional theranostic platform in oncology and facilitate significant advancements in this research-active area. Refereed/Peer-reviewed

This paper addresses a differential game with three players: a target, an attacker and a defender, where the Attacker aims to capture the Target while avoiding being captured by the Defender, and the Defender tries to defend the Target from being captured by the Attacker while trying to capture the Attacker. There are two coupled pursuit-evasion problems in this game: Attacker-Target and Defender-Attacker. Firstly, a game of kind in this differential game is considered and a so-called barrier is constructed to divide the space into three disjoint regions: the win region of the Attacker, the win region of the team of the Target and Defender and the uncertain region in which neither of three players can win the game. Secondly, a game of degree is considered by introducing a payoff function. Some explicit expressions of optimal strategies for the players are obtained. Finally, the numerical solutions of optimal strategies and the corresponding optimal trajectories for three players with different initial conditions are provided.

Speech emotion recognition is a challenging and significant task. On the one hand, the emotion features need to be robust enough to capture the emotion information, and while on the other, machine learning algorithms need to be insensitive to model the utterance. In this paper, we presented a novel framework of speech emotion recognition to address the two above-mentioned challenges. Relative Entropy based Normalization (REN) was proposed to normalize the supervectors of Gaussian Mixture Model-Universal Background Model (GMM-UBM) as the features to emotions. The Kernel Extreme Learning Machine (KELM) was adopted as the classifier to identify the emotion represented by the normalized supervectors. Experimental results on the EMR_1309 corpus showed the proposed framework outperformed the state-of-the-art i-vector based systems.

We report an approach to make two-terminal antenna-coupled AlGaN/GaN high-electron-mobility-transistor self-mixing terahertz detectors. Fluorine ion implantation is used to increase the threshold voltage of the AlGaN/GaN two-dimensional electron gas. An optimal implantation dose can be reached so that the detector responsivity is maximized at zero gate voltage or with the gate floating. The relationship between the ion dosage and the threshold voltage, electron mobility, electron density, responsivity, and noise-equivalent power (NEP) is obtained. A minimum optical NEP of 47 p W / Hz is achieved from a two-terminal detector at 0.65 THz. The capability of two-terminal operation allows for the design of a large array of antenna-coupled high-electron-mobility transistor detectors without the demanding needs of routing negative gate voltage lines around the antenna array and minimizing the gate leakage current.

The responsivity and the noise of a detector determine the sensitivity. Thermal energy usually affects both the responsivity and the noise spectral density. In this work, the noise characteristics and responsivity of an antenna-coupled AlGaN/GaN high-electron-mobility-transistor (HEMT) terahertz detector are evaluated at temperatures elevated from 300 K to 473 K. Noise spectrum measurement and a simultaneous measurement of the source–drain conductance and the terahertz photocurrent allow for detailed analysis of the electrical characteristics, the photoresponse, and the noise behavior. The responsivity is reduced from 59 mA/W to 11 mA/W by increasing the detector temperature from 300 K to 473 K. However, the noise spectral density maintains rather constantly around 1–2 pA/Hz1/2 at temperatures below 448 K, above which the noise spectrum abruptly shifts from Johnson-noise type into flicker-noise type and the noise density is increased up to one order of magnitude. The noise-equivalent power (NEP) is increased from 22 pW/Hz1/2 at 300 K to 60 pW/Hz1/2 at 448 K mainly due to the reduction in mobility. Above 448 K, the NEP is increased up to 1000 pW/Hz1/2 due to the strongly enhanced noise. The sensitivity can be recovered by cooling the detector back to room temperature.

Dynamic Music Emotion Prediction is crucial to the emerging applications of music retrieval and recommendation. Considering the influence of temporal context and hierarchical structure on emotion in music, we propose a Deep Bidirectional Long Short-Term Memory (DBLSTM) based multi-scale regression method. In this method, a post-processing component is utilised for individual DBSLTM output to further enhance the ability of temporal context processing and a fusion component is to integrate the output of all DBLSTM models with different scales. In addition, we investigate how the difference of sequence length between the training and predicting phase affects the performance of DBLSTM. We conduct our experiments on a public database of Emotion in Music task at MediaEval 2015, and the result shows that our method achieves significant improvement when compared with the state-of-art methods.

Dynamic music emotion prediction is to recognize the continuous emotion contained in music, and has various applications. In recent years, dynamic music emotion recognition is widely studied, while the inside structure of the emotion in music remains unclear. We conduct a data observation based on the database provided by Free Music Archive (FMA), and find that emotion dynamic shows different properties under different scales. According to the data observation, we propose a new method, Double-scale Support Vector Regression (DS-SVR), to dynamically recognize the music emotion. The new method decouples two scales of emotion dynamics apart, and recognizes them separately. We apply the DS-SVR to MediaEval 2015, Emotion in Music database, and achieve an outstanding performance, significantly better than the baseline provided by organizer.

Music Dynamic Emotion Prediction is a challenging and significant task. In this paper, We adopt the dimensional valence-arousal (V-A) emotion model to represent the dynamic emotion in music. Considering the high context correlation among the music feature sequence and the advantage of Bidirectional Long Short-Term Memory (BLSTM) in capturing sequence information, we propose a multi-scale approach, Deep BLSTM (DBLSTM) based multi-scale regression and fusion with Extreme Learning Machine (ELM), to predict the V-A values in music. We achieved the best performance on the database of Emotion in Music task in MediaEval 2015 compared with other submitted results. The experimental results demonstrated the effectiveness of our novel proposed multi-scale DBLSTM-ELM model.

The urban heat island is generally conducted based on ground observations of air temperature and remotely sensing of land surface temperature (LST). Satellite remotely sensed LST has the advantages of global coverage and consistent periodicity, which overcomes the weakness of ground observations related to sparse distributions and costs. For human related studies and urban climatology, canopy layer urban heat island (CUHI) based on air temperatures is extremely important. This study has employed remote sensing methodology to produce monthly CUHI climatology maps during the period 2000–2013, revealing the spatiotemporal characteristics of daytime and nighttime CUHI during this period of rapid urbanization in Shanghai. Using stepwise linear regression, daytime and nighttime air temperatures at the four overpass times of Terra/Aqua were estimated based on time series of Terra/Aqua-MODIS LST and other auxiliary variables including enhanced vegetation index, normalized difference water index, solar zenith angle and distance to coast. The validation results indicate that the models produced an accuracy of 1.6–2.6 °C RMSE for the four overpass times of Terra/Aqua. The models based on Terra LST showed higher accuracy than those based on Aqua LST, and nighttime air temperature estimation had higher accuracy than daytime. The seasonal analysis shows daytime CUHI is strongest in summer and weakest in winter, while nighttime CUHI is weakest in summer and strongest in autumn. The annual mean daytime CUHI during 2000–2013 is 1.0 and 2.2 °C for Terra and Aqua overpass, respectively. The annual mean nighttime CUHI is about 1.0 °C for both Terra and Aqua overpass. The resultant CUHI climatology maps provide a spatiotemporal quantification of CUHI with emphasis on temperature gradients. This study has provided information of relevance to urban planners and environmental managers for assessing and monitoring urban thermal environments which are constantly being altered by natural and anthropogenic influences.

Single electron transistors (SETs) are known to be extremely sensitive electrometers owing to their high charge sensitivity. In this work, we report the design, fabrication, and characterization of a silicon-on-insulator-based SET scanning probe. The fabricated SET is located about 10 μm away from the probe tip. The SET with a quantum dot of about 70 nm in diameter exhibits an obvious Coulomb blockade effect measured at 4.1 K. The Coulomb blockade energy is about 18 meV, and the charge sensitivity is in the order of 10-5-10-3 e/Hz1/2. This SET scanning probe can be used to map charge distribution and sense dynamic charge fluctuation in nanodevices or circuits under test, realizing high sensitivity and high spatial resolution charge detection.