Search

"C Programming Design" is an important professional basic course for students to opens the door to the intelligent world in the future. The teaching case takes the teaching process of "Understanding C Programming" as the carrier, designs a modern digital technology teaching case based on the OBE idea with the students as the center and the whole process of academic evaluation as the link. The design of the teaching case integrates online and offline teaching, builds a new five-in-one teaching system integrating education idea, teaching goal, teaching resource, teaching method and teaching evaluation, so as to achieve the goal of improving teaching efficiency and talent development quality. Finally, the implementation effect of the teaching case and the problems existed are analyzed, and the continuous improvement measures are proposed.

Summary Room-temperature phosphorescence (RTP) has received extensive attention in recent years due to its various potential applications. Pure organic RTP materials may overcome the shortcomings of harsh preparation conditions and high toxicity associated with metal complexes and rare-earth-containing materials. Although crystallization strategies can yield rigid environments isolated from oxygen, its inferior processability can significantly limit the corresponding practical applications. Considering their flexibility, stretchability, processability, and thermal stability with flexible long chains and intertwined structures, polymers are an ideal material for achieving RTP and realizing practical applications. In this review, we summarize the latest progress in polymer-based RTP materials, elaborating on two strategies of chemical bonding and physical blending. In the last part of the review, challenges and perspectives of amorphous polymers are proposed to improve RTP performance and develop new applications.

Some experimental phenomena in theoretical mechanics are not suited for realization in an ordinary laboratory. Virtual experiments in theoretical mechanics can be realized using Maple, a scientific computing software. Although most previous studies have only drawn static images of the trajectories of objects, this thesis investigates the use of Maple’s animation capabilities to achieve dynamic simulations of theoretical mechanics experiments. The benefits of dynamic simulation are that the trajectory of the object over time can be seen and the motion characteristics of the object, such as the direction of approach of the incoming track, can be fully understood. The dynamic simulation of theoretical mechanics experiments will fully mobilize students’ interest in learning, transform abstract physical models into visual physical images and deepen their understanding of physical phenomena. In addition, virtual experiments enable students to train their programming skills and improve their scientific research literacy.

Recently, deep learning-based methods have reached an excellent performance on License Plate (LP) detection and recognition tasks. However, it is still challenging to build a robust model for Chinese LPs since there are not enough large and representative datasets. In this work, we propose a new dataset named Chinese Road Plate Dataset (CRPD) that contains multi-objective Chinese LP images as a supplement to the existing public benchmarks. The images are mainly captured with electronic monitoring systems with detailed annotations. To our knowledge, CRPD is the largest public multi-objective Chinese LP dataset with annotations of vertices. With CRPD, a unified detection and recognition network with high efficiency is presented as the baseline. The network is end-to-end trainable with totally real-time inference efficiency (30 fps with 640p). The experiments on several public benchmarks demonstrate that our method has reached competitive performance. The code and dataset will be publicly available at https://github.com/yxgong0/CRPD.

To achieve high coverage of target boxes, a normal strategy of conventional one-stage anchor-based detectors is to utilize multiple priors at each spatial position, especially in scene text detection tasks. In this work, we present a simple and intuitive method for multi-oriented text detection where each location of feature maps only associates with one reference box. The idea is inspired from the twostage R-CNN framework that can estimate the location of objects with any shape by using learned proposals. The aim of our method is to integrate this mechanism into a onestage detector and employ the learned anchor which is obtained through a regression operation to replace the original one into the final predictions. Based on RetinaNet, our method achieves competitive performances on several public benchmarks with a totally real-time efficiency (26:5fps at 800p), which surpasses all of anchor-based scene text detectors. In addition, with less attention on anchor design, we believe our method is easy to be applied on other analogous detection tasks. The code will publicly available at https://github.com/xhzdeng/stela.

Recently, scene text recognition methods based on deep learning have sprung up in computer vision area. The existing methods achieved great performances, but the recognition of irregular text is still challenging due to the various shapes and distorted patterns. Consider that at the time of reading words in the real world, normally we will not rectify it in our mind but adjust our focus and visual fields. Similarly, through utilizing deformable convolutional layers whose geometric structures are adjustable, we present an enhanced recognition network without the steps of rectification to deal with irregular text in this work. A number of experiments have been applied, where the results on public benchmarks demonstrate the effectiveness of our proposed components and shows that our method has reached satisfactory performances. The code will be publicly available at https://github.com/Alpaca07/dtr soon.

Based on the dynamics angle, the paper finds a suitable launch window based on the pork-chop diagram, and then uses the conic splicing method to carry out the preliminary design of the orbit. The two-body model is used to design the orbit at each stage, and finally under the precise mechanical model, using differential correction method, accurate track design based on B plane parameters.

This paper systematically studies the application of orbital dynamics theory in spacecraft orbit design. In this paper, two linearized equations with J2 relative term perturbation are introduced firstly, and their advantages and disadvantages are compared and analyzed. On this basis, an improved method is proposed and applied to the design of hovering orbit and fast flying orbit, and good results are obtained in some cases. The traditional numerical solution algorithm for ballistic design is innovated and improved, and the improved Newton iterative method is comprehensively applied to ensure that the algorithm can still converge quickly under the condition of inaccurate initial value; The simulation results show that the orbit integrated rapid design method studied in this paper is reliable, fast and accurate, and can be applied to the rapid orbit design of spacecraft.

Complex space missions have higher and higher requirements for orbit representation, orbit prediction and orbit control. Traditional orbit parameters have many shortcomings in orbit representation and orbit extrapolation. From the point of view of improving the shortcomings of traditional orbital parameters, this paper analyzes several groups of orbital parameters including unit quaternions. Based on a group of quaternion orbital elements composed of unit quaternion and vector parameters such as radial distance, radial velocity and angular momentum, the orbital dynamics and kinematics equations are derived, and the transformation relationship between quaternion orbital elements and traditional orbital parameters is given. The simulation example proves the effectiveness of improving the traditional kinematic equation method of orbit parameters.

Recently, methods based on deep learning have dominated the field of text recognition. With a large number of training data, most of them can achieve the state-of-the-art performances. However, it is hard to harvest and label sufficient text sequence images from the real scenes. To mitigate this issue, several methods to synthesize text sequence images were proposed, yet they usually need complicated preceding or follow-up steps. In this work, we present a method which is able to generate infinite training data without any auxiliary pre/post-process. We tackle the generation task as an image-to-image translation one and utilize conditional adversarial networks to produce realistic text sequence images in the light of the semantic ones. Some evaluation metrics are involved to assess our method and the results demonstrate that the caliber of the data is satisfactory. The code and dataset will be publicly available soon.

Previous approaches for scene text detection usually rely on manually defined sliding windows. This work presents an intuitive two-stage region-based method to detect multi-oriented text without any prior knowledge regarding the textual shape. In the first stage, we estimate the possible locations of text instances by detecting and linking corners instead of shifting a set of default anchors. The quadrilateral proposals are geometry adaptive, which allows our method to cope with various text aspect ratios and orientations. In the second stage, we design a new pooling layer named Dual-RoI Pooling which embeds data augmentation inside the region-wise subnetwork for more robust classification and regression over these proposals. Experimental results on public benchmarks confirm that the proposed method is capable of achieving comparable performance with state-of-the-art methods. The code is publicly available at https://github.com/xhzdeng/crpn .