Search

The path signature feature (PSF) which was initially introduced in rough paths theory as a branch of stochastic analysis, has recently been successfully applied to the field of pattern recognition for extracting sufficient quantity of information contained in a finite trajectory, but with potentially high dimension. In this paper, we propose a variation of path signature representation, namely the dyadic path signature feature (D-PSF), to fully characterize the trajectory using a hierarchical structure to solve the rotation-free online handwritten character recognition (OLHCR) problem. We adopt the deep neural network (DNN) as classifier, and investigate three hanging normalization methods to improve the robustness of the DNN to rotational distortions. Extensive experiments on digits, English letters, and Chinese radicals demonstrated that the proposed D-PSF, jointly with hanging normalization and DNN, achieved very promising results for rotated OLHCR, significantly outperforming previous methods.

This paper develops and experimentally validates a 3D-printed snake robot prototype. Its structure is designed to allocate limited room for each functional module (including an external power module, battery power module, the wireless control and transmission module and some detective sensors), so as to ensure the snake robot works in different environments. In order to control and track the snake robot, a low-cost MEMS-IMU (micro-electro-mechanical systems inertial measurement unit)-based snake robot motion tracking system is developed. Three algorithms (low-pass filter, baseline calibration, and Kalman filter) are used to eliminate noise from IMU’s acceleration data, thus minimizing the noise influence to tracking accuracy. Through signal processing, the IMU acceleration data can be effectively used for motion tracking. The result from the video tracking software is employed as a reference for comparison, so as to evaluate the motion tracking algorithm efficiency. The comparison results demonstrate high efficiency of the proposed IMU-based motion tracking algorithm.

With increasing emissions of industrial wastewater and poor control measures, environmental pollution has become a serious issue haunting China’s economic development. Meanwhile, the current pollution management policy system in China is mainly under the supervision of the central government and executed by local governments. Under the current economic growth model of China, the industrial sector remains the dominant segment of our economy, which makes the Total Factor Efficiency (TFE) evaluation and policy analysis of industrial wastewater control decisive factors concerning China’s future economic growth and sustainable development. Based on existing studies of China and abroad, and with the help of a Data Envelope Analysis (DEA) model, this paper used 39 industrial sectors and their input-output data from 2003 to 2014 of China as Decision Making Units to calculate the TFE of wastewater control in different industrial sectors of China. Moreover, we have designed and adopted our own MATLAB programming for optimization solutions of multi-variable constrained nonlinear functions in order to obtain a more accurate estimation of the TFE of wastewater control. Based on our calculation results, this paper further explained the difference in TFE and policy implications across typical industries in China, and offered policy recommendations accordingly.

A 3D printed snakebot prototype is developed and experimentally validated in this paper. The structure is designed to allocate limited room for each functional module (including an external power module, battery power module, the wireless control & transmission module and some detective sensors) to ensure the snakebot work in different environments. For controlling and tracking the snakebot, a MEMS IMU (Micro-Electro-Mechanical-Systems Inertial Measurement Units) based snakebot motion tracking system is developed. To minimize the noise influence to tracking accuracy, three algorithms (low-pass filter, baseline calibration, and Kalman filter) are used to eliminate noise from IMU's acceleration data. Through signal processing, the IMU acceleration data can be effectively used for motion tracking. To evaluate the motion tracking algorithm efficiency, the result from a real-time video tracking is employed as a reference for comparison. The comparison results demonstrate high efficiency of the proposed IMU based motion tracking algorithm.

One of the serious issues that China faces during its fast economic development is the low input–output efficiency of water and energy resources and growing water pollution. With the current economic development model of China, economic growth still requires large input of water resource and energy resource. This paper has focused on the total factor efficiency of water resource and energy resource by each province in China. We treat the undesirable outputs as outputs in the DEA-SBM Model instead of as inputs in previous studies, and design a new MATLAB programming to achieve optimization solutions of multi-variable constrained nonlinear functions to evaluate the Total Factor Efficiency of Water resource (TFEW) and the Total Factor Efficiency of Energy (TFEE) in China accurately. By using the method, this paper has analyzed the TFEW and TFEE in China from 2003 to 2014 by economic zones and typical provinces and provided corresponding policy recommendations.

As a developing country with insufficient water resources, China's water environment management and performance evaluation have important research value. The three provinces (Henan, Hubei, and Hunan) in central China with typical significance in geographical location and water resources governance were selected as research objects in this paper. Based on the principal component analysis (PCA) method and the pressure-state-response (PSR) model, a comprehensive evaluation system for the water environment in those three provinces during 2011&ndash, 2017 was established in this paper. The evaluation results show that: (1) The water environment management and performance evaluation of the three provinces in central China were generally poor in 2011&ndash, 2012, but the overall trend was rising, (2) in 2013&ndash, 2014, the situation was improved compared to the previous two years, but needed further enhancement, (3) in 2015&ndash, 2017, the water environment management and performance of the three provinces showed significant improvement. Among them, the Hubei Province had the highest water environment evaluation value (1.692), and the Henan Province had the most significant progress (from 0.043 to 1.671). The contributions of this paper are: (1) The comprehensive evaluation model based on PCA and the PSR model was constructed to analyze the sustainable development of water environment in central China, (2) the performance evaluation system for water environment management, which could comprehensively evaluate the performance of water environment treatment and effectively reveal the correlation between various indicators, was established. The principal factors in water environment management can be obtained by this evaluation system. Based on the analysis of the reasons underlying the above changes, the corresponding policy recommendations for improving water environment management and performance in central China were suggested in order to provide a reference for further improvement of water environment management in developing countries.

This paper presents an investigation of several techniques that increase the accuracy of online handwritten Chinese character recognition (HCCR). We propose a new training strategy named DropDistortion to train a deep convolutional neural network (DCNN) with distorted samples. DropDistortion gradually lowers the degree of character distortion during training, which allows the DCNN to better generalize. Path signature is used to extract effective features for online characters. Further improvement is achieved by employing spatial stochastic max-pooling as a method of feature map distortion and model averaging. Experiments were carried out on three publicly available datasets, namely CASIA-OLHWDB 1.0, CASIA-OLHWDB 1.1, and the ICDAR2013 online HCCR competition dataset. The proposed techniques yield state-of-the-art recognition accuracies of 97.67%, 97.30%, and 97.99%, respectively., Comment: 10 pages, 7 figures

After decades of economic development in China, its energy consumption still relies heavily on coal, and the dominant segment of our economy&mdash, the industrial sectors with large coal input&mdash, have caused massive wastewater pollution. To provide a decision-making basis for the sustainable development of Chinese industry and economy, this paper selected 39 industrial sectors in China and studied their Total Factor Efficiency (TFE) using Dual Target Variables of coal input and wastewater control during 2003&ndash, 2014. The results showed that 82.05% of the total industries within our study scope had an efficiency level below 0.5. Over the entire study period of 2003&ndash, 2014, only the national monopoly industries and high and new tech industries were able to maintain the optimal efficiency level. In comparison with the existing Data Envelope Analysis (DEA) Model and studies on TFE using multi-target variables methods, this paper has probably made two improvements. First, we improved the original DEA model to calculate the TFE of two target variables (coal input and wastewater control) at the same time. Second, we developed an algorithm to calculate the TFE of industrial coal input and wastewater control from 2003 to 2014, based on the official input-output statistics of China.

Rationale:The corpus callosum, which is the most important fiber pathway linking the bilateral hemispheres, plays a key role in information access, as well as the functional coordination and reorganization between the bilateral hemispheres. However, whether the corpus callosum will undergo s

Carrier-Smoothed-Code (CSC) can effectively suppress the multipath effect and receiver noise in the code-based differential positioning, which is widely adopted by the Ground Based Augmentation System (GBAS). However, the accuracy of smoothed code can be degraded significantly by the Code-Carrier-Divergence (CCD) effect. The CCD monitoring technology of for BeiDou Satellite System (BDS) is still under developing. This paper analyzed and compared two methods of CCD monitoring in GPS and BDS, respectively. Based on the CCD monitoring results, it was found that the tow-order linear invariant filter is better for CCD in terms of detection time and estimation accuracy. With the given constant false alarm rate, CCD detection threshold can be obtained. Furthermore, the sigma-inflation method is used to compensate the non-perfect distribution of the constructed test statistics. The real-world experiment of GPS and BeiDou is implemented to make a comparative analysis. The experimental result shows that the two-order detection method is more applicable to BDS due to the better availability.

Inspired by the theory of Leitners learning box from the field of psychology, we propose DropSample, a new method for training deep convolutional neural networks (DCNNs), and apply it to large-scale online handwritten Chinese character recognition (HCCR). According to the principle of DropSample, each training sample is associated with a quota function that is dynamically adjusted on the basis of the classification confidence given by the DCNN softmax output. After a learning iteration, samples with low confidence will have a higher probability of being selected as training data in the next iteration; in contrast, well-trained and well-recognized samples with very high confidence will have a lower probability of being involved in the next training iteration and can be gradually eliminated. As a result, the learning process becomes more efficient as it progresses. Furthermore, we investigate the use of domain-specific knowledge to enhance the performance of DCNN by adding a domain knowledge layer before the traditional CNN. By adopting DropSample together with different types of domain-specific knowledge, the accuracy of HCCR can be improved efficiently. Experiments on the CASIA-OLHDWB 1.0, CASIA-OLHWDB 1.1, and ICDAR 2013 online HCCR competition datasets yield outstanding recognition rates of 97.33%, 97.06%, and 97.51% respectively, all of which are significantly better than the previous best results reported in the literature., 18 pages, 8 figures, 5 tables

Due to the CCD array and scenery relative motion, the image quality would be degraded. In order to get the relationship between the CCD array move quantity and the image quality, the corresponding degraded images are achieved by changing the CCD array velocity, pitch, roll and yaw variation. Then, the appropriate image quality evaluation method to research the sensitivity of images degraded is required. The results show that the velocity CCD array is the main factor affecting the degradation of the image, and the velocity increase, the image quality degraded sharply, but the motion value has more than 6 pixels in image plane, evaluation values is unchanged basically, the quality of image is extra negative.

Carrier-Smoothed-Code (CSC) can effectively suppress the multipath effect and receive noise in the code based differential positioning, which is widely adopted by the Ground Based Augmentation System (GBAS). Since the BeiDou Navigation Satellite System (BDS) code measurement suffers from the severer measurement noise, the CSC is necessary for the BDS based GBAS. However, the accuracy of smoothing pseudorange can be degraded significantly by the ionospheric delay error, which caused by the Code-Carrier Divergence (CCD) effect. The theoretical analysis shows that the CCD will greatly affect the precision of the smoothed pseudorange, thus, ionospheric delay error has to be considered for the services related to the safety-of-life (SoL). Based on the traditional CCD monitoring techniques, this paper proposed a new two first-order linear time invariant filter sequentially to improve the CCD performance. Meanwhile, sigma-inflation method is used to obtain the CCD detection threshold considering the impact from the non-ideal distribution of the constructed statistic. The real-world experiment result has shown that the proposed method can suppress the measurement noise more effectively when compared with the traditional first-order linear monitoring method with satisfying the predefined false alarm requirement, which make the proposed method more adaptable to the small bias from CCD.

Low energy efficiency and severe environmental pollution are two growing issues haunting China’s fast economic development. Under China’s current development model, economic growth still heavily depends on massive energy input, which makes energy efficiency one crucial topic in order to secure future sustainable development of China. This paper focuses on the unique energy ownership structure of China, and designs and adopts MATLAB programming for optimization solutions of multi-variable constrained nonlinear functions to obtain results that can better reflect China’s energy efficiency and solutions to sustainable development of China. Using this model, this paper conducts an empirical analysis on the impact of difference energy investment behaviors between China’s state-owned entities and non-state-owned business on China’s Total Factor Energy Efficiency (TFEE) and sustainable development from 2003 to 2014. We find that Beijing and Shanghai represent the highest energy efficiency level in China. However, except for the more developed regions in Eastern China, for other provinces in Northeast China, Central China and Western China, the TFEE of energy investment by state-owned and non-state-owned economies are both quite low compared with the optimal level. Based on the above findings from the empirical study and detailed analysis by region, this paper discusses the possible reasons for China’s low TFEE and provides implications and policy recommendations.

In view of current broadcast station problems existed in the bus, bus automatic voice broadcast station system based on RFID is designed in this paper. The system can realize function of automatic broadcast station by combining single chip microcomputer technology with digital speech and wireless communication technology. The overall scheme, hardware and software design of the system are given in the paper, the system can lower the driver’s labor intensity, increase bus security, it has lower cost and higher practical value. The system is suitable for the development of intelligent traffic and the construction of intelligent city.

The line control for multiple intersection is a hot spot in the area of traffic control,solving signal lights timing in main road plays extremely essential role in alleviating traffic jam. The paper introduces a compound control method to three intersection on the main road. Through gathering the sensor data, analyzing traffic situation, the system can choose one of control methods to deal with the real traffic light in synchronous fuzzy control and asynchronous fuzzy control methods according to real traffic situation. The simulate procedure have been written with matlab, as the contrast, under the same environment and parameter, synchronous fuzzy control and asynchronous fuzzy control procedure are also been written. The result indicates the compound control method can effectively reduce the vehicle average delay and improve the traffic conditions in intersection.

Speech recognition is to study so that the machine can accurately hear the voice of the contents of people's problems, that is, accurate identification of words. Since the 90's it has been the focus of speech recognition has shifted to large vocabulary, continuous speech, speaker-up, man-machine voice communication has become a research focus. Use one kinds of arithmetic to study speech recognition: Dynamic Time Warping (DTW). Build a basic test platform of speech recognition using DTW.

Nowadays, the doorbell style is varied and many modern bells are electronic products but most of their functions are single. In this paper, it will show a design method of the digital doorbell using Protel99 SE software. Its mainly controller is AT89C52 SCM and its input device is 4×4 keyboard, DS1302 clock circuit and AT24C02 storage circuit and ISD1420 voice circuit are its intermediate response equipments, and playback speech circuit as well as HD44780 LCD display circuit as its intermediate response equipments. The working status of this digital doorbell can be set by users themselves, besides a common electronic bell function; it also has some more new functions. Through the time of keying and different music has been set up, the doorbell user can easily distinguish different types of visitors. In addition, the doorbell can also provide some information to the visitors, in the form of voice or text. What's more, the user can query the information every time, to make sure if there are visitors or not. In a word, the doorbell is a humanized design.