A kind of face snap machine and apply the monitoring system of this kind of face snap machine
Technical field
This utility model belongs to safety monitoring technical field, is specifically related to the face snap machine of a kind of automatic candid photograph face picture and applies the monitoring system of this kind of face snap machine.
Background technology
Along with the development of safety monitoring industry Yu face recognition technology, more and more back-end servers matched with CCTV camera are provided with face identification device.But as shown in Figure 1, CCTV camera is as front end node, it does not have CPU and memory element, i.e. CCTV camera does not have data processing function, and it must process at video information transmission to back-end server, for saving bandwidth, the data of its transmission are the most compressed video flowing, and single frames picture distortion in compressed video flowing is serious, definition is low, this picture is utilized to be difficult to clearly observe face, it is also difficult to be carried out the recognition of face of high accuracy by technological means.Further, since the video flowing of transmission contains substantial amounts of insignificant data, the such as information data such as clothes, environment, not only occupy transmission bandwidth, but also occupy memory space, improve hardware cost.Therefore, there is that a kind of new face snap machine of necessary design, to improve the accuracy of recognition of face, and reduce memory space, the hardware requirement of transmission data bandwidth and cost.
Summary of the invention
The purpose of this utility model is to propose a kind of face snap machine and apply the monitoring system of this kind of face snap machine, to improve the accuracy of recognition of face, and reduces memory space, the hardware requirement of transmission data bandwidth and cost.
Face snap machine of the present utility model is by CPU, taking module, for carrying out recognition of face from the photo captured by taking module and face picture being carried out identification and the memory module stored, constitute for carrying out the communication module of communication with the server of rear end, described CPU respectively with taking module, identify and memory module, communication module is connected and controls taking module, identify and memory module, the work of communication module, the data output end of described taking module is connected with the data input pin of identification and memory module, the data output end of described identification and memory module is connected with the data input pin of communication module.
The monitoring system applying above-mentioned face snap machine is become with described face snap mechanism by the server being furnished with monitor screen, the communication module of described face snap machine is connected with server communication, store with the face picture transmission that face snap machine is obtained and process to server, described monitor screen is connected with server, the face picture information received with display server.
Above-mentioned face snap machine is as front end node, not only it is configured with taking module and communication module that tradition CCTV camera must have, it is provided with CPU and identification and memory module, therefore this face snap machine oneself has just possessed data-handling capacity, independently the photo captured by taking module can be carried out recognition of face, and the face picture identified and select out is stored, finally utilize communication module to transmit to the server of rear end.Captured by above-mentioned taking module for photo rather than video flowing, and uncompressed processes, and therefore identifies and memory module just can utilize the photo that definition is high to carry out recognition of face, improve the accuracy of identification.When with back-end server communication, the data of face snap machine transmission are the human face photo through selection rather than video flowing, therefore substantially reduce data volume, thus reduce the requirement to communication bandwidth and corresponding hardware cost.Above-mentioned CPU, taking module, identification and memory module, communication module are the prior art used in current monitoring system, originally the CPU being deployed on back-end server, identification and memory module have simply been deployed on the face snap machine of front end by this utility model, hardware, the software of CPU, identification and memory module are not made improvement.
Further, described server is provided with or connects warning devices, when the similarity of the human face photo so prestored in received human face photo with server exceedes predetermined threshold, server just can utilize warning devices point out and report to the police, it is achieved automatic monitoring function.
Further, described server is computer or mobile phone, and when server is mobile phone, each Security Personnel may be by mobile phone and synchronizes to receive human face photo information so that monitoring work is more flexible.Mobile communication network based on current 3G, 4G, mobile phone is quickly collected picture data and has been fully able to realize.
Face snap machine of the present utility model and apply the monitoring system of this kind of face snap machine mainly the face snap machine of front end node to be improved, face snap machine is made to have had certain data-handling capacity, i.e. taken photo is carried out recognition of face and corresponding photo storage in front end, the most it is not only back-end processing and is provided previously by effective material, without again carrying out the large overhead computings such as video analysis, decrease the waste of memory space and the hardware requirement to communication bandwidth.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of the monitoring system of prior art.
Fig. 2 is the schematic diagram of monitoring system of the present utility model.
Detailed description of the invention
Below against accompanying drawing, by the description to embodiment, detailed description of the invention of the present utility model is described in further detail such as the mutual alignment between the involved shape of each component, structure, each several part and annexation, the effect of each several part and operation principle etc..
Embodiment 1:
As shown in Figure 2, the face snap machine of the present embodiment is by CPU, taking module, for carrying out recognition of face from the photo captured by taking module and face picture being carried out identification and the memory module stored, constitute for carrying out the communication module of communication with the server of rear end, described CPU respectively with taking module, identify and memory module, communication module is connected and controls taking module, identify and memory module, the work of communication module, the data output end of described taking module is connected with the data input pin of identification and memory module, the data output end of described identification and memory module is connected with the data input pin of communication module.
In above-mentioned face snap machine, taking module is preferably with the camera that can shoot with high-speed and continuous, and shooting speed is more preferably greater than 20 frames per second, to reach not omit the purpose of face.Communication module can be wired or wireless communication mode,
The operation principle of face snap machine is as follows: the image in target scene is shot by first taking module, and be sent to identify and memory module by taken photo (without compression) in real time, identify and whether memory module detection photo exists face, if be detected that face, then the picture of human face region is stored, finally, the picture of these human face regions can be transferred on server by communication module in the way of active push or passive extraction according to configuration.Above-mentioned process is controlled by CPU, and corresponding control program is common program, and here is omitted.
The monitoring system applying above-mentioned face snap machine is become with described face snap mechanism by the server being furnished with monitor screen, the communication module of described face snap machine is connected with server communication, store with the face picture transmission that face snap machine is obtained and process to server, described monitor screen is connected with server, the face picture information received with display server.
Further, described server connects warning devices, when the similarity of the human face photo so prestored in received human face photo with server exceedes predetermined threshold, server just can utilize warning devices point out and report to the police, it is achieved automatic monitoring function.
Further, described server is mobile phone, and each Security Personnel may be by mobile phone and synchronizes to receive human face photo information so that monitoring work is more flexible.Mobile communication network based on current 3G, 4G, mobile phone is quickly collected picture data and has been fully able to realize, and the data processing function of mobile phone can serve as server the most completely.
Above-mentioned face snap machine is as front end node, not only it is configured with taking module and communication module that tradition CCTV camera must have, it is provided with CPU and identification and memory module, therefore this face snap machine oneself has just possessed data-handling capacity, independently the photo captured by taking module can be carried out recognition of face, and the face picture identified and select out is stored, finally utilize communication module to transmit to the server of rear end.Captured by above-mentioned taking module for photo rather than video flowing, and uncompressed processes, and therefore identifies and memory module just can utilize the photo that definition is high to carry out recognition of face, improve the accuracy of identification.When with back-end server communication, the data of face snap machine transmission are the human face photo through selection rather than video flowing, therefore substantially reduce data volume, thus reduce the requirement to communication bandwidth and corresponding hardware cost.Above-mentioned CPU, taking module, identification and memory module, communication module are the prior art used in current monitoring system, originally the CPU being deployed on back-end server, identification and memory module have simply been deployed on the face snap machine of front end by this utility model, hardware, the software of CPU, identification and memory module are not made improvement.