Aiming at the problems of complex modeling process, high cost and long cycle, this paper proposes a new 3D imaging algorithm based on the segmentation of key areas of images. By using the edge detection of key target areas of images, combined with image correction and perspective transformation, the key areas of two-dimensional images are converted into three-dimensional images through the fusion of segmented images, so that the images can form three-dimensional naked eye 3D simulation effect of the senses. The algorithm firstly gets the target image and the background image through the segmentation of the key areas of the image,and then processes the target image, corrects it through perspective transformation, and then migrates the corrected image, carries on the grayscale processing, and finally overlays and fuses to achieve the purpose of thickening the image, and forms the three-dimensional effect on the vision. Experimental results show that this method can process images with different complexity, and can produce stereoscopic effects on both single object and multiple objects.

In order to solve the problem that the old people and children are not easy to distinguish the garbage classification method,this paper presents a set of automatic reminder and inquiry device based on human body induction and voice prompt. The system ismainly composed of STM32 mcu core control system, LD3320 speech recognition module, human body induction module, speech broadcast module and TFT display screen. The automatic garbage classification prompt is realized by integrating speech word baseand matching algorithm in the system, and the acousto-optic prompt is set at the corresponding garbage input port. Through the garbage entry test, the functions of garbage classification and automatic reminder are achieved.

In view of the problems existing in the talent training of electrical information specialty in our university, the "1235" talent training mode is proposed. The theoretical teaching and practical teaching are reformed by adjusting the curriculum system, reforming the teaching methods, building teaching teams, encouraging students to participate in workshops and various competitions, school and enterprise integration and other means to improve the students' application ability.

This paper deeply explores on the construction method of the new engineering department and plans the training goal of excellent network engineer. Based on the platform of Huawei ICT College, this paper implements the training programs in the form of curriculum placement, training of outstanding teachers and cooperation between schools and enterprises in the development of practical courses, puts forward the innovative ideas of undergraduate professional cognition practice and production practice, and introduces the training mode of our university in the process of implementation.

Based on the design of the AGV robot in the Internet of Things, this paper focuses on the research on its steering system,power system, communication system, control system and other key systems, and uses simulation software to design and analyze the stress of the main supporting parts of theAGV robot.

Deep learning is a hot topic in computer vision and natural language processing. In many applications, the performance of deep neural networks (DNN) is better than traditional methods, and has been successfully applied to tasks such as modulation classification and radio signal representation learning. In recent years, studies have found that deep neural networks are vulnerable to adversarial attacks and lack robustness to "adversarial examples". Regarding the robustness of the neural network communication signal recognition algorithm, the author regards the data that has been attacked by the PGD as model-based data, and inputs the data into the neural network, which makes the signal recognition and classification results wrong. Then with the help of model-based defense algorithms, which are robust training algorithm and adversarial training algorithm, the experimental results after training show that both methods have good defense effects.

Although the traditional conditional random fields (CRF) method can accommodate any length of context information and the feature design is flexible, but the training cost is high and the model complexity is high, especially in the sequence tagging task,because the joint probability distribution of the whole tagging sequence needs to be calculated, its shortcomings are more exposed.For this reason, this paper combines a structured support vector machine (SSVM) method to do part of speech tagging research according to Agglutinating Language word formation features and context information of corpus. Compared with traditional SVM, this model can fit the distribution of feature functions by adding additional constraints, and then be used to deal with tagging in different fields. In this paper, the Agglutinating Language part of speech tagging experiment results show that the accuracy of SSVM is higher compared with the traditional part of speech tagging algorithm.

With the rapid expansion of the meteorological station network in Fujian province, the types of instruments and equipment are constantly updated. However, the informatization construction of meteorological stations is relatively backward with a lack of scientific and effective system platform for management and maintenance, which greatly increases the complexity of the work. In this paper, the basic geographic information data and meteorological station metadata of Fujian province is utilized to construct the information management system with the B/S architecture. The functions are designed and implemented, such as information visual query of meteorological station network as well as the examination and approval management of station migration, establishment and evacuation. The developed system can provide a unified service of network information for provincial, municipal and county business personnel and effectively improve the management ability of meteorological station network information.

With the development of informationization, more and more courses are arranged in the computer room. It makes the demand of teachers and students to the computer room real-time remote storage increasing urgent. This paper introduces the open source software ownCloud to build file cloud storage for computer room teaching, which is helpful to realize real-time remote storage function for teachers and students in school. This paper also describes the basic network environment and server environment of file cloud storage in detail, as well as the method of building ownCloud file storage in the LNMPenvironment and its application effect.

Coding is like the ID card of a product, which can provide accurate and comprehensive information for data management personnel and facilitate follow-up tracking and information inquiry of products, while the traditional manual management mode has disadvantages such as low efficiency, poor confidentiality and tedious inquiry. Therefore, the use of PowerBuilder language to develop a professional software for electric submersible pump product code management, can not only avoid human error caused by chaos in the formulation of product code, but also greatly improve work efficiency and accuracy, while facilitating management, maintenance and retrieval.

In current training trade, online training has become more and more popular, especially in Shanghai Pudong Software Park, many IT companies want to train their employee by online mode. Therefore the skill training center of Shanghai Pudong Soft- ware Park developed an online training system. In order to satisfy the high volume of the request, the DB Service System of this sys- tem uses the MyCAT component to store the data in splitted DB. And in order to improve the efficiency, the DB Service System uses the Redis sentinel cluster to cache the data. After the release, the system works well, and provides the solid support for the online training system.

To deal with the issue of lifecycle security of Publicly Posted Composite Documents (PPCD) which was proposed in re- cent years, in this paper, we propose a Secure Self-Destruction Scheme for Publicly Posted Composite Documents with Hierarchical Identity-based Encryption (HICDSS), which can protect a user’s sensitive PPCD automatically destructed after a period of time. We form a new structure of the PPCD with different security levels. We firstly encrypt the different security documents into ciphertexts by different symmetric keys and encrypt the symmetric keys using the HIBE algorithm; then use the Mingle Shares Generaion algo- rithm to make the part of document’s ciphertext and the symmetric key’s ciphertext into mixed ciphertext, and sent the mixed ci- phertext into the Object Storage System; finally we analyze and prove the security, efficiency and performance of the HICDSS scheme.

An intelligent medicine box is designed with STC15F2K60S2 as the core to realize the functions of timing voice prompt, voice prompt for the elderly entering the range of taking medicine, automatic recognition and control of taking medicine, and it can change the timing time and the content of each recording in real time. By using single-chip computer programming control, the sys- tem is divided into self-checking module, timing setting module, near-body recognition module, drug-taking recognition module, LCD module, voice recording / playback control module. It can realize the time of speech warning and taking medicine for the elder- ly, especially the elderly with Alzheimer's disease, to automatically recognize that the elderly have come to the medicine box and tak- en the medicine by voice prompt, to automatically recognize whether the medicine has been taken out and the empty medicine box has been put back.

Aiming at the problems of poor real-time performance, long processing time, and slow resource request of the existing big data processing platform, this paper uses Storm real-time computing technology, combined with Flume, Kafka, Zookeeper and other big data processing components to design a real-time data processing platform. It uses tornado+WSGI+Apache technology to build a Web server, and uses Echarts technology to visually analyze the processing results. This article uses the website access log as the da- ta source to verify the platform. Through the test, the platform can complete the real-time calculation of the website's click-through rate and the number of visitors. It has the characteristics of stability, reliability and simple operation.

Compared with MOOC, SPOC has the advantages of improving teaching efficiency, suitable for blended teaching, and im- proving teaching analysis efficiency, which makes it able to deeply integrate with traditional offline courses. However, traditional of- fline courses have problems such as fixed teaching time, tight class hours, and untimely teaching feedback. Therefore, this article proposes a SPOC-based hybrid teaching model for higher vocational computer majors, and points out that the construction of SPOC platform requires the construction of corresponding teaching resource modules, online test module, discussion area and learning anal- ysis module, and discusses the specific development of online and offline teaching mode.

COVID-19 breaks the traditional teacher-centered and teacher-filled teaching mode. In the epidemic period, more colleges and universities adopt the online and offline teaching mode to teach. This paper analyzes the advantages of Superstar Learning Plat- form, and uses this platform as an auxiliary teaching platform to carry out mixed teaching for Computer Application Foundation, aiming at exploring a new teaching mode suitable for learners' needs in the practice of mixed teaching, and providing reference for future teaching.

In today's society, with rapid development of information, network security is becoming more and more important. The de- mand for network security talents continues to grow. The computer network specialty in technical schools pays more and more atten- tion to the training of network security talents. However, there are some problems in the teaching process of Network Security. This paper mainly discusses on the problems encountered in the teaching of network security and the countermeasures.

Combining with teaching practice, this article explains the environment and development background of the integration of production and education, points out the problems and shortcomings of the integration of production and education training model, explores the UI talent training model, and practices the talent training ideas and teaching methods based on integration of production and education. It also puts forward the executable practical models from the aspects of teaching mode, training methods and teacher improvement, which provides a useful reference for the teaching and personnel education of computer majors and other similar high- er vocational majors.

Facing the increasingly severe employment problems today, secondary vocational students need to have good professional knowledge skills and practical experience. This article introduces the construction scenario from teaching innovation and skill im- provement combined with current employment status for the Internet of Things major in our school. It helps students to understand their job responsibilities in the course in advance, enhance employment confidence, consolidate professional knowledge, improve professional comprehensive ability, and has important practical significance for secondary vocational students.