Welcome to Tencent Cloud + community, get more Tencent mass technology practice dry goods oh ~
This article was published in the Cloud + Community column by Yunplus Community Technology Salon
Speaker: Wu Chen, R&d Director of Sichuan Yunjian Technology Development Co., LTD. With more than 10 years of r&d team management experience, I organized and participated in the r&d of many software products and the construction of ministry and provincial information platform. In 2017, it was selected into the test and testing expert database of highway and water transportation engineering and participated in the compilation of related industry management specifications.
Good afternoon, everyone. I would like to communicate with you about the application practice of smart site performance attendance system. This lecture is divided into six parts: frontier, product analysis, system architecture, main technologies, functional analysis and application prospects. There must be a specific reason for making a product: to strengthen the application of four new technologies in response to the call of highway quality engineering construction of the Ministry of Transport. Zhaozhou Bridge and Dujiangyan, which we learned in primary school textbooks, including today’s National Aquatics Center, Water Cube and bird’s nest, are all typical quality projects. In the process of quality, the application of four new technologies should be strengthened, including the application of new materials, new equipment, new technology and new process.
Performance attendance management system belongs to the application of intelligent site, the purpose is to strengthen the management of performance personnel. The performance attendance management system mainly uses Tencent cloud AI technology to achieve intelligent performance attendance management. We use attendance very much, but on the basis of performance attendance, combined with the actual situation of the site to achieve intelligent. Performance attendance brings us several questions: What is performance attendance? Second, why is attendance attendance required? Third, how to perform attendance? And how to achieve intelligent performance attendance. Attendance check is to check the attendance, through this way to obtain the situation of employees or team in a specific place, a specific period of time, performance broadly speaking is to fulfill the agreement, for both parties in the contract is to exercise their own rights and obligations. Performance in engineering projects often refers to the performance of the construction party. Performance attendance is actually a part of our staff attendance management.
Second, why is attendance attendance required? Performance attendance is an important part of contract performance management. Strengthening personnel performance management can protect project construction and quality. In practical application, the object of performance attendance refers to some important personnel including project manager, chief engineer, responsible person of safety production, resident design representative of design unit, test and testing engineer, etc. Personnel performance management should also meet the requirements of documents related to national highway and water transport projects. Impostor can be avoided by checking the performance of personnel.
We analyzed some mainstream attendance devices before developing performance attendance. Mainstream attendance, such as fingerprint attendance is the most widely used method, which has the advantage of good identification effect and fast speed. However, some personnel cannot use it. The second ID card attendance, the application is also very wide, the advantage is fast, but there are employees on behalf of the phenomenon. Some companies have developed a device that gives attendance information to workers wearing hard hats as they walk by. But there is also the problem of punching in.
The third paper card clock attendance. There is also a phenomenon of clocking in. The fourth is face recognition attendance machine, there are restrictions on the use of the environment, can not be used outdoors. The focus of our research is to realize both fixed attendance and mobile attendance. Of course, there is also iris recognition, iris recognition equipment in the domestic production of less. It is very expensive and is only used in places with high security requirements. There are also finger vein identification, which is mainly produced by Japanese manufacturers. The disadvantage is that the equipment identification speed is slow, not suitable for large-scale use. But it can solve some of the problems of clocking in. These are some of the mainstream attendance devices. Others are internet-based attendance checks. Such as nail nail attendance, including enterprise wechat and so on. These are based on account number or mobile phone number verification, which can record who is in attendance and where, but do not solve the impostor problem.
We often pay attention to the authenticity of personnel when we use performance attendance in engineering projects. It is necessary to develop a new attendance system because the construction party and the construction party belong to the relationship between supervision and supervision, and the management department has higher requirements for the authenticity of the information of the performance personnel. Moreover, the construction project site environment is mostly harsh, so the cloud-based core function is particularly important. We now add cloud functions by using Tencent cloud’s artificial intelligence technology. Personnel change is common in construction site. Because of intelligent means, changing information is automatically identified. It also facilitates deployment. Artificial intelligence, face recognition and mobile Internet attendance together form the intelligent attendance.
We will implement intelligent attendance from six aspects: First, ensure that it is cloud-based. In this way, the deployment cost is low, management and maintenance is convenient, and data authenticity is ensured.
The second is based on artificial intelligence. Tencent cloud artificial intelligence, face recognition technology including face detection and analysis, positioning of five senses, face comparison witness, in vivo detection, etc. The third is based on cloud computing. Through the construction of cloud platform to achieve terminal equipment, data processing and resource sharing. Fourth, connect big data. On the one hand, the production of attendance data in the construction process is realized. On the other hand, the generated data can be collected, sorted and reused to provide decision-making basis for the management. The same applies to regulatory agencies, government authorities.
Fifth access to mobile applications, including smart phones, tablets, mobile apps. Sixth support for intelligent devices, including face recognition, front-end punch, data upload. Through these six aspects to ensure our intelligent attendance. Intelligent performance attendance has the following characteristics: It can collect and input basic facial information of personnel; Automatic recognition, collection and comparison based on human face information; Support personnel information and personnel id card information verification; Support the setting of attendance location and judgment of abnormal attendance location; Support the early warning of abnormal attendance information of key posts and absence information of key posts; Implement online management of personnel information changes, and verify the attendance results on the cloud.
Ensure the reliability of data through the initial test of the client and the final test of the cloud. The system architecture adopts the “cloud + terminal” structure design, including PC terminal management and Tencent cloud service, supporting multi-terminal mobile attendance and fixed point attendance, supporting multiple projects access and attendance architecture. The main technology of intelligent attendance mainly includes face comparison, certification of nuclear body; In vivo detection, cloud SMS LBS; GPS and so on. Tencent cloud face comparison is based on facial features to calculate the similarity of two faces, automatic identification. After comparing the face comparison functions of Alibaba, Baidu and Huawei, we still choose Tencent Cloud.
In the system, it is judged to be the same person by comparing the photos of personnel attendance with those of verification. Before and after the judgment and in vivo detection together to ensure that each attendance information is true and reliable, to avoid impostor. Using a mobile phone to compare one photo with another without the addition of a biopsy can also overlap. So we added such a technology as liveness detection to the attendance system. The picture above explains that when two photos are 75% or 80% alike, they belong to the same person. This is the contrast.
Second, use the function of the human body. The verification body of Tencent cloud is also the core body of human face. It verifies the user’s identity through the user’s self-portrait video or self-portrait and another user’s pre-retained photo. This part is mainly to help improve the efficiency of business management and reduce labor costs. Because the project is widely dispersed, each person needs the construction party, the management party to carry out a verification is very tedious. At the same time, on-site verification of personnel change is also a tedious process. So we use the technology of human identification to keep the data moving and the people on the road less.
Third, use in vivo detection. Tencent cloud in vivo detection is through the face feature point positioning tracking recognition of 3D face reconstruction model, judge whether it is a real person. Supports multiple platforms, including CPU and GPU computing modes, and flexible deployment.
Fourth, LBS positioning service is applied. The location service also belongs to Tencent Cloud, which is a location-based service that obtains the location information of mobile terminal users, including geographic coordinates or geodetic coordinates, through the radio communication network of telecom and mobile operators, GSM and CDMA networks. With the support of geographic information system platform, value-added services are provided for users. The range of attendance can be determined by setting the geographical location of fixed attendance points and determining the effective range of attendance on mobile terminals. The current location can be automatically obtained after determining the attendance range. When the valid range can be registered and automatically alert.
Other cloud host, Open CV and other technologies. After 2017, a large number of Technical services of Tencent Cloud have been adopted. As long as a machine has been put on record, other servers can also be put on record quickly. The deployment of servers and application deployment can be basically done within half an hour. Tencent cloud SMS also has its own characteristics, support signature Settings. Users can set up, set up after passing the audit can start to use. The third one is the Open CV technology, which carries out preliminary judgment and verification on mobile terminals.
Cloud inspection intelligent performance attendance system uses advanced face recognition technology, through the combination and application of face recognition network attendance machine, smart phone APP and Web management background, to realize the integration of authentication, to achieve the performance attendance of key positions. At the same time, it can also realize dynamic analysis of attendance data, including the real-time warning function of abnormal data of missing positions. Function points mainly include the following aspects:
The first is a personal face recognition attendance machine. Face recognition attendance machine it mainly plays the role of front-end data collection. The main functions include personnel information registration, face attendance and data upload; The second mobile APP is mainly used for face attendance and identity verification; The third performance attendance management cloud platform includes attendance management, approval management, statistical analysis, system Settings, etc. The fourth API mainly provides the interface for mobile APP to call application programs, and finally uses the technical services of Tencent cloud and artificial intelligence such as face comparison and human identification.
Face recognition attendance machine is mainly for fixed point personnel attendance, data uploaded to the cloud for authenticity verification. Mobile attendance adds artificial intelligence technology to the verification. Of course, it is the personnel login by the PC side first input personnel information, the user through the mobile phone number to log in the mobile phone. Did not verify the body first to carry out a nuclear body, after the nuclear body can enter the punch card interface. So when you clock in for face recognition, if you identify the process including your photo or abnormal position, there will be corresponding records to remind.
At the same time, users can also complete their own clock records, including abnormal attendance information query. Leave application and other related management. First, an employee’s name, ID number and other identity information will be registered in the cloud. For the first time, you need to check in first. After checking in, you can enter the check in interface and punch in after success. Attendance management cloud platform includes attendance management; Approval management, performance management and system management.
The sixth aspect is application prospect. Face recognition solutions are widely used in many industries. With the development of technology, market expansion, face recognition technology in real life more and more application scenarios, value and role is becoming more and more. Face recognition can be used in construction sites to play a supervisory role in the safety management of the site environment and construction personnel; The attendance of construction supervision personnel can be queried at any time, and witness sampling can also be done.
Sometimes there will be construction units and supervision units together to cheat the owners of the situation. Even if the supervisors do not sample on site, they can input the corresponding information in the office. In fact, they can also use the face recognition technology and location positioning technology of Tencent Cloud. Face recognition first identifies the corresponding personnel, and can obtain sampling place and sampling time through positioning technology. The same applies to supervisors.
For all in and out of the site construction personnel, especially important equipment including tower cranes, elevators and other equipment to achieve face verification operation, to ensure that each work has a trace to follow. In order to avoid the project supervision lax generation of safety risks can be through the face verification technology, the relevant personnel in the use of tower crane or elevator and other important equipment, must be witness, so as to ensure that I must be in the operation. Of course, my application outlook is these, and I hope we can explore more solutions together. Tencent cloud ARTIFICIAL intelligence will make life better, thank you.
Q&A
**Q: ** Thank you for your speech. You said just now that a positioning system is required for clocking in. Now many mobile phones can simulate the location, and many mobile phones have a software that can change the location to a certain place. Is there any preventive mechanism for such a system?
**A: ** This is A contradictory relationship. Every technical solution cannot be airtight. However, this application is mainly aimed at our construction site. We also consider the problems arising from the development of technology, but this should not be the resistance to our application.
**Q: ** Has the solution you proposed been used in the construction site?
**A: ** This project started in March 2008, and actually the construction period was only two months. It was launched in Chengdu in May.
**Q: ** When it comes to application, the input of this cost is to ask the supervisor to supervise it on site, because it is not to set up a face recognition machine? If you’re building roads the machinery and the cameras have to be like a project, they have to be moved, they have to be invested. Another aspect is that the quality of workers is not high. Will this approach affect the enthusiasm of workers? Can this work efficiency be guaranteed?
**A: ** Due to this situation, we adopt A solution such as cloud + terminal. First, we have A fixed attendance point. For example, this is installed in the project department, each bidding section has a project department. For each project, there are many work areas, and people in each work area cannot punch in the project department every day. In view of this situation, mobile solutions are provided, that is, the corresponding software is installed on the mobile phone, which can also achieve fixed punch in. Therefore, there will be no such situation as you mentioned. The location of the project department is fixed. At the same time, we also support mobile terminal, for some people who need mobile office, they can punch in their mobile phone.
It also has no effect on work motivation.
Question and answer
Language requirements for AI development?
reading
Safety reports | Internet malicious crawler analysis: during the first half of 2018 from a panoramic perspective the crawler and the crawler
Safety reports | SSH brute force trend: from cloud to iot device migration
Install a panel for your CVM.
This article has been published by Tencent Cloud + community authorized by the author.Cloud.tencent.com/developer/a…
Welcome to Tencent Cloud + community or follow the wechat public account (QcloudCommunity), the first time to get more mass technology practice dry goods oh ~
Massive technical practice experience, all in the cloud plus community!