Facial expression recognition is an important branch of facial recognition in the field of computer vision. Influenced by many factors including the diversity of facial expressions, changes in head posture and the environment, it is a big challenge to the work of facial expression recognition. Focused on traditional convolutional neural network, owning to its large number of parame- ters and the limited improvement in facial expression recognition accuracy compared with applying traditional machine learning al- gorithms, we proposes an improved convolutional neural network model based on the depthwise separable convolution structure. We make the experiment on the Fer2013 gray expression recognition dataset based on our model, the result shows that the network struc- ture of the model is optimized, the number of model parameters is greatly reduced, and the utilization efficiency of model parame- ters is high, while ensuring a high accuracy rate of 68.31% compared to the traditional convolutional neural network.

Aiming at the inefficiency of the traditional parking fee management method which relies on manual statistical records, a parking lot management system based on Alibaba Cloud Computing OCR technology is designed. The main functions include li- cense plate recognition, parking fee management, parking information query and payment information query, etc. This system uses Shiro security framework to verify the user's identity, uses SSM Web project framework to design, uses Java language to program, and uses MySQL database to manage the background data.

At present, elderly care has become a hard problem as the aged tendency of the population is deepening. So designing a care system is necessary for the elderly people. Moreover, with the popularity of mobile devices and the gradual maturity of 5G tech- nology, the system gets essential technical support for the realization. This article elaborates on the design and implementation of an elderly care system. This system utilizes Android sensors and Baidu Map API map positioning technology to realize the functions of fall alarm, real-time positioning, sports health information recommendation as well as convenient operation.

The rise of Internet of things technology, wireless communication and computer technology in recent years has attracted the scientific community's attention to wireless sensor networks, and the study of reasonable distribution coverage of detection areas. It is also necessary to maximize coverage.Artificial bee colony algorithm is a kind of optimization method which imitates the bee be- havior. It can obtain more superior convergence results for unconstrained numerical optimization problems. Because the existence of artificial bee colony algorithm is easy to be limited to the local optimal solution, the process of the intermediate stagnation problem, the need for a longer search time, an improved artificial bee colony algorithm is proposed, which can speed up the convergence speed in the later stage. The improved artificial bee colony algorithm can effectively reduce the redundancy and prolong the lifetime of the sensor network by optimizing the node coverage.

Performance test has become an important part of software test and an effective way to find software performance prob- lems. In view of the problems existing in the traditional performance test teaching process, the experimental teaching content and the experimental teaching method are reformed. For the experimental teaching content, the basic experimental teaching and the compre- hensive experimental teaching are designed. The experimental teaching introduces the method of combination of theory and experi- ment and project-driven method, and finally achieves good results.

Production-oriented Approach is a novel teaching model motivated by ability aims, concentrating the coordination between teaching resource input and ability-aimed output, emphasizing practices and capacity development. Production-oriented Approach has been extensively studied in this paper and employed to reconstruct the course of Web Service and RESTful Technology based on the feature analysis on the software engineering professional abilities. The empirical data manifests the enhancements in learning ini- tiative, practical ability fostering and the utilization of education resources.

Learning analysis technology is a new technology widely used to promote learning. It can perform quantitative structural analysis on learning data, combine artificial intelligence, expert system and teacher system structure, build a bridge between the data structure and meaning understanding, enrich the application of data information, and effectively carry out analysis and understand- ing. This article firstly describes the connotations, connections, and characteristics of learning analysis technology and deep learn- ing; then puts forward the exploration and research on deep learning strategies for higher vocational expansion students under the background of learning analysis technology and combining deep learning related theories.

It is important for the supply side to reform in education to promote the transformation of local universities into applica- tion-oriented universities. The establishment of a scientific, professional and dynamic adjustment mechanism is an important driving force for the application-oriented transformation and development of local undergraduate universities. This paper analyzes the prob- lems existing in the major settings of local undergraduate colleges and universities by making academic discipline plans, and estab- lishing professional evaluation mechanism, enrollment professional dynamic control mechanism, discipline professional develop- ment and protection mechanism，so as to construct a stable professional dynamic adjustment mechanism, optimize the structure of discipline and specialty, improve the effective supply of applied talents, and enhance the core competitiveness of universities.

Aiming at the low efficiency of manual screening resume, the lack of timely feedback of results and the high cost, and com- bining with the demand of "cloud recruitment" for efficient resume screening in the epidemic period, an online resume information screening platform is developed. The platform adopts MVC mode, uses servlet framework, javabean for data storage processing, JSP display and AHP filter mode. The platform has the advantages of high filtering efficiency, quick feedback, low cost, simple opera- tion. It can be used in the scene of enterprise cloud recruitment, college online enrollment, talent evaluation and so on.

This paper analyzes the problems existing in the teaching of data structure course, and puts forward the reform from the as- pects of strengthening basic programming, visualized explanation of abstract theory, combination of C++, Java programming, ratio- nal use of online platform, and stratification of exercises, aiming at improving students' practical ability, learning interest and self- confidence.

With the development of smart detoxification, the demand for intelligent safety precautions for detoxification is more and more urgent, especially for the fine control of detoxification personnel. The intelligent control of personnel is one of the important manifestations of security informationization in detoxification places. Taking the Compulsory Quarantine Detoxification Center in Lituo as an example, this paper analyzes the problems faced by intelligent personnel control and puts forward corresponding counter- measures.

Aiming at the single teaching method of the course of C Language Programming in traditional teaching and easily causing two-level differentiation of students' learning effect, the 1 + e teaching mode combining offline one classroom teaching with on- line e-network teaching is built. This paper takes students from different classes of the computer major of our school as the research object, and compares the traditional teaching model and the 1 + e teaching model based on academic performance and program- ming ability. Practice shows that the 1 + e fusion teaching mode is more suitable for the modern teaching of the C language pro- gramming course, and the teaching effect is good.

As an emerging network structure, the capsule network uses vector output instead of scalar output, which can capture the spatial relationship between image features and improve the limitations of convolutional neural network. This paper firstly trains the capsule network to achieve image classification, obtains the predictive label of the image, determines the category of the query im- age, and then uses the feature parameters in the digital capsule layer of the network as the feature vector of the image. The feature vector is used to find images similar to the query image in the category set of the query image. In this paper, experiments are carried out on the FASHION-MNIST and CIFAR10 datasets respectively. The experimental results show that the proposed method can bet- ter extract the features of the images and obtain good image retrieval results.

Based on the main contents of meteorological information network construction, integrated network architecture and net- work security provided by Qingdao Meteorological Bureau for the 2018 SCO summit, this paper introduces the meteorological net- work construction program of the Qingdao Meteorological Bureau. The Intelligent Resilient Framework (IRF) method is used to re- build the core network architecture; meanwhile, the network security equipment is also increased. The network security policies are set up. The network construction program has effectively improved the operational performance of Qingdao meteorological informa- tion network and played an important role of meteorological service during the SCO summit. At the same time, the implementation of this network program has also laid a good foundation for the Qingdao meteorological information network system optimization and upgrade.

In recent years, the scale and speed of network development in our country is amazing, and the popularity is more and more wide, so the IPv4 network protocol cannot meet the needs of address space and network security. The emergence of new gener- ation network protocol IPv6 solves these problems. In this paper, IPv6 network protocol is studied, and its network settings and secu- rity solutions in smart campus are analyzed.

Existing resource recommendation algorithms make resource recommendation based on user preferences, without consider- ing the subjective reluctance of users to personal privacy or the data mining of interests and hobbies, or the risk of privacy disclosure of third-party servers. To solve the problem of user privacy, this paper puts forward the K-anonymous privacy protection algorithm. After the generalization of sensitive attributes of user query requests, it constructs a logical anonymous query request equivalence class, uses the method of data rotary to enable users in the same equivalence class to randomly forward the received data from other users. Because the privacy attributes that each person wants to protect are different, the weight of its sensitive attribute is different, so this paper proposes the weight summation formula based on the sensitive attribute combined with the sensitive attribute weight value set by the user independently, and recommends the optimal selection scheme for the platform user. The security analysis shows that this method can effectively resist similarity attack, background knowledge attack and capture server attack. Experiments show that this method not only satisfies the correctness of matching results, but also enhances the privacy protection performance in the process of resource recommendation.

With the development of Internet information technology in China, the construction industry has begun to abandon the tra- ditional manual drawing, calculation and design methods for the design of three-dimensional views or models of certain projects, us- ing BIM software technologies such as Revit, ArchiCAD, and VectorWorks. The construction of three-dimensional terrain scenes such as roads, bridges, and water Conservancy will be carried out to promote the digital and integrated development of large-scale engineering projects and improve the work efficiency and work quality of the design and construction of construction projects.

This calibration device of radio comprehensive tester is studied and established in this paper. It solves the verification and calibration of a large number of general-purpose radio comprehensive testers. The calibration device consists of four calibration sub- systems and calibration software system, including RF generator, audio generator, audio analyzer and RF analyzer. It also researches on the configuration of standard equipment and analyzes the uncertainty of the calibration devices, ensuring accurate and reliable val- ue transfer.

For the defect that the mechanical properties of polypyrrole supercapacitors decrease with charge and discharge cycles, and the cycle stability is poor, a compound method of polypyrrole electrode for supercapacitors based on multi-walled carbon nanotubes is proposed. The structure of the formed polypyrrole coated multi-walled carbon nanotubes effectively increase the specific surface area of the electrode material, the utilization rate of the active material and the electrical conductivity, improve the rapid charge and discharge performance of the electrode material, and greatly improve cyclic stability of polypyrrole. The composite electrode materi- al of polypyrrole and functionalized multi-walled carbon nanotubes for supercapacitors is prepared by pulse current deposition meth- od. It is scanned at a scanning rate of 1000mV?s -1 in a 3mol?L -1 potassium chloride electrolyte, after 100,000-cycle, the capacity on- ly declines by 16%.

According to the content of the course of Communication Equipment Operation and Maintenance , this paper matches it with the professional qualifications, adopts the design idea of combining virtual simulation platform with real hardware equipment operation, and the educational mechanism of combining work with study and cooperation between school and enterprise. Taking the operation and maintenance of TD-SCDMA network equipment as an example, the teaching design and practice based on "task-driv- en" and integration of "teaching and doing" are carried out, which improve the learning effect of students and the teaching level.

in the information age, the course of Computer Application Basics plays a fundamental and instrumental role, but there are a few researches on the learning needs of students in the existing literatures. Based on the classification of students with different learning experiences, the article studies on their learning needs for the course content. The results confirm that the correlation be- tween the learning needs of the four types of students with different basic levels and the current course content settings gradually de- creases. Aiming at the problems reflected in the research process, such as, the current curriculum adopts the non-discriminatory teaching in a unitary standard, which makes it impossible to achieve the teaching effect that "teachers teach students according to their aptitude and students take what they need", it puts forward improvement suggestions for reference.