Search

Generally, medical images are usually characterized by complex background and low resolution. Also, text's location is sometimes random. Therefore, traditional Optical Character Recognition (OCR) technology cannot be well applied to recognize medical images which contain many factors. Aiming at this problem, a character recognition algorithm for medical images is proposed in this paper. The Efficient and Accurate Scene Text (EAST) scene detection algorithm is used in text detection and the Convolutional Recurrent Neural Network (CRNN) model is used in character recognition. The problem of low recognition rate that caused by the low resolution will be solved by adding a multi-scale convolution kernel to the model. The designed algorithm is tested by Medpix datasets, and the experimental results show that the proposed method has a higher precision and a higher recall than the traditional recognition methods.

Adaptive feedback cancellation (AFC) method is widely adopted for the purpose of reducing the adverse effects of acoustic feedback on the sound reinforcement systems. However, since the existence of forward path results in the correlation between the source signal and the feedback signal, the source signal is mistakenly considered as the feedback signal to be eliminated by adaptive filter when it is colored, which leads to a inaccurate prediction of the acoustic feedback signal. In order to solve this problem, prediction error method is introduced in this paper to remove the correlation between the source signal and the feedback signal. Aiming at the dilemma of Modified Least Mean Square (MLMS) algorithm in choosing between prediction speed and prediction accuracy, an improved MLMS algorithm with a variable step-size scheme is proposed. Simulation examples are applied to show that the proposed algorithm can obtain more accurate prediction of acoustic feedback signal in a shorter time than the MLMS algorithm.

The quality and cost of coke are directly affected by the proportion of different categories of coal, while the coal blending in coking process is a complex one with multiple objectives and constraints. In this study, a multi-objective simulated annealing algorithm (MOSA)-based coal blending optimization approach in coking process is proposed. The objective function which considers the coke quality indexes and its cost for coal blending is constructed, along with the corresponding constraints. Then, a modified MOSA based on the decision space search strategy is presented to calculate the optimal solution, in which the value of the variable in intermediate solution will mutate with a certain probability during the search process, thereby expanding the search range. The validation experiments using actual coal blending data are carried out. The results indicate that the strategy proposed in this paper is capable of searching the Pareto-optimal (PO) solutions accurately and comprehensively, which exhibits better performance than the existing ones, and can provide guidance for coking production.

The switch machine is a vital device for operating railway turnouts and plays an important role in safe running of train. Abnormal working state of switch machine will affect safety of railway transportation. Therefore, it is necessary to monitor the working state of switch machine timely. The switch machine gap is an important indication of its working state, due to the complicated external environment, the traditional methods and the image-based methods for monitoring switch machine gap are hard to satisfy the application demands. To solve this problem, this paper proposes a new detection algorithm for switch machine gap. Firstly, the raw image is downsampled and the detection region of the gap is determined by matching the characteristic of the detection column and indication rod in the downsampled image. Secondly, multiple image processing techniques and the RANSAC (Random Sample Consultation) algorithm are used to obtain the gap’s pixel distance in the detection region. Finally, the gap’s real distance is calculated by the mathematical relationship between its pixel distance and its real distance. Based on the proposed algorithm, a monitoring controller for the switch machine gap is designed and evaluated on the ZD(J)9 switch machine. The experimental results show that the performance of the monitoring controller satisfies the application requirements, and the switch machine gap can be monitored in real-time and accurately, which makes the railway transportation more reliable and safety.

Train broadcast controller is an indispensable part of modern train design, which is simple, practical and fully functional. Most of the traditional train broadcast controllers are designed based on finite state machines. The system designed in this way is generally simple in function and poor in reliability. With the continuous development of urban rail transit, passengers are more and more demanding for train broadcasting. Therefore, this paper introduces the design of train broadcast controller based on embedded Linux platform and hierarchical state machine. Firstly, several functions of the train broadcast controller are determined from the user's point of view. Secondly, according to the different implementation scenarios and steps of each function, the super-state, sub-state and the transition conditions between states are determined. Finally, the train broadcast controller is designed and implemented based on UML diagram of hierarchical state machine.

Recently, vision-based target positioning technology has received considerable attention in the field of computer vision. The positioning technology has many advantages such as faster speed, higher accuracy and more stable and reliable positioning results. Therefore, it plays an important role in the fields of robot technology, military reconnaissance, geographic survey and field measurement. Based on the positioning technology, the paper designs an embedded vision system with DM816x microprocessor (the processing module) and CCD camera equipped with an infrared filter (the acquisition module), which realizes the recognition and positioning of the fire. Moreover, with 200 groups of experiments in a warehouses, the fire detection system has an alarm accuracy rate over 98%, and positioning accuracy of fire is far higher than the national standard of China. In addition, an automatic fire protection system is proposed, which consists of the fire detector and the water cannon, then it can automatically extinguish the fire when a fire is detected.

Good air-conditioning system is an important reflection of passenger comfort. Traditional air-conditioning controller interface is mostly realized by the combination of PLC(Programmable Logic Controller) and configuration software. However, with the continuous development of urban trains, the requirements of control accuracy, response speed and equipment expansion are constantly improved. Therefore, this paper presents the design and development of train air-conditioning controller interface based on Linux platform and Qt framework. Firstly, the design method and requirements of the interface are determined from the user-centered point of view, and then on this basis the framework of the interface is determined. Secondly, the interface drawing, control association, data refreshing and thread safety which are realized by Qt framework and C++ design pattern are introduced. Thirdly, the PID(Proportion Integral Derivative) control model of air-conditioning controller is established with temperature as input and air-conditioning unit as control object.

In light of significant complexity of the byproduct gas system in steel industry (which limits an ability to establish its physics-based model), this study proposes a data-based predictive optimization (DPO) method to carry out real-time adjusting for the gas system. Two stages of the method, namely the prediction modeling and real-time optimization, are involved. At the prediction stage, the states of the optimized objectives, the consumption of the outsourcing natural gas and oil, the power generation and the tank levels, are forecasted based on a proposed mixed Gaussian kernel-based prediction intervals (PIs) construction model. The Jacobian matrix of this model is represented by a kernel matrix through derivation, which greatly facilitates the subsequent calculation. At the second stage, a rolling optimization based on a mathematical programming technique involving continuous and integer decision-making variables is developed via the prediction intervals.

Due to the data diversity and complexity in industrial system, the accuracy of data-based modeling might be largely affected by such a series of issues. Aiming at the energy system in steel industry, this study proposes a fuzzy modeling based on Gaussian membership expression. First, in the stage of sample selection, the industrial data set is divided into a number of clusters, from which the representative sample are chosen based on a variable step rule. Second, given the industrial data usually accompany with high level noise and anomaly points, a fuzzy modeling based on Gaussian membership is proposed, where a sample reliability coefficient is introduced to alleviate the negative impact produced by ill-posed data, and the model parameters solution is explicitly derived later. The proposed method has been applied to the practice of gas flow prediction in a steel plant. To verify its performance, a number of experiments are conducted by using the data coming from the energy center in the plant. The results indicate that the proposed method greatly improves the prediction accuracy and efficiency, which plays a significant role in data-based modeling for the industrial system.

currently, most LED controllers for the metro carriage display are driven directly by I/0 of MCU and the display interface is very simple, therefore it's difficult to meet the multiform requirements of video terminal. In order to meet this requirement, this paper designed the software system of new type metro display LED controller based on Ethernet, analyzed its function and achieved an implementation. This new type metro display LED controller takes Linux and Qt system as a developing platform and communicated with TCMS in process data and message data through the Ethernet. This new type metro display LED controller has been applied to a driverless metro and the result shows that its response speed is fast and its display effect tends to be diversiform.

The accuracy of quaternion method in strap-down inertial navigation used for posture recognition is subjected to sensor data drift. Analysis shows that quaternion is a state variable in discrete time linear constant system. A method based on Kalman filter to correct the error is proposed. Recursive calculations on state variables can effectively reduce system error. Experiments show that this method significantly increases the accuracy of posture recognition.

The Train Communication Network (TCN) is a high-performance local area network for information transfer of modern train control, state detection, fault diagnosis and passenger service and so on. The Multifunction Vehicle Bus (MVB) is a critical part of the TCN and the design of the MVB's protocol analysis tool plays a very important function. Supported by the friendly Windows graphics user interface, a powerful MVB's protocol analysis tool is developed in this paper and this tool fills in a vacancy of the technical application presently. Starting from the MVB protocol, the frame construction of the MVB protocol analyzer and the function design of each module of up level computer software are introduced. Finally this protocol analyzer is applied to the developing and debugging process of state detection, fault diagnosis and vehicle equipment of the TCN.

Along with the development of train communication network, Train-Borne Announcement System Controller (TASC) has become an essential part of the underground and city rail vehicle. According to this requirement and based on the multifunction vehicle bus standard, this paper presents the general design, hardware buildup and software implementation of the TASC. The TASC has perfect functions, transmits data fast and correctly, and satisfies the real time requirement of the train control and manage system. The TASC has applied successfully to a line of Guangzhou underground and runs stably.

A Feedback Control method of Deadline Assignment (FC-DA) is presented to reduce output jitter of control tasks. The prevailing PID control law is employed, whose parameters can be adaptively adjusted online according to CPU load and jitter. The schedulability of whole task set is guaranteed dynamically. A simulation experiment shows the improvement when compared with proportional-only law.