CN203894772U

CN203894772U - Mass face detecting and identifying system

Info

Publication number: CN203894772U
Application number: CN201420107900.1U
Authority: CN
Inventors: 夏东林; 骆名洲; 夏霆
Original assignee: SUZHOU ZHONGANSHI TECHNOLOGY Co Ltd
Current assignee: SUZHOU ZHONGANSHI TECHNOLOGY Co Ltd
Priority date: 2014-03-11
Filing date: 2014-03-11
Publication date: 2014-10-22
Anticipated expiration: 2024-03-11

Abstract

The utility model discloses a mass face detecting and identifying system including a front end device and a background processing device connected with each other. The front end device includes a continuous video stream camera and a snapshot camera connected with an identification unit provided with a face identifying and cutting module and a first network transmission module. The background processing device includes a front-end processor, an operation table and a storage device connected with the server, wherein the front-end processor is provided with a second network transmission module corresponding to the first network transmission module. According to the utility model, captured images are identified through the identification unit. After human faces are detected in the images, the parts of the human faces are cut from the captured images and are then sent to a remote server for comparison and identification. Therefore, the amount of wireless transmission data is reduced, the transmission rate is increased and mass identification is realized.

Description

A kind of magnanimity people's face is found and recognition system

Technical field

The utility model relates to electronic information communication technology, and a kind of magnanimity people's face relating in particular to is found and recognition system.

Background technology

In prior art, modal face identification system has: human face identification work-attendance checking system, face recognition door control system, offender's photo Compare System.

Wherein, immediate with the present invention is offender's photo Compare System, this is the Compare System of a set of basic static, it is to adopt the photo of preserving complete wanted criminal in database, when running into suspect object, it is taken pictures, by the photo of the photo of shooting and maintenance is directly contrasted.

The problem of this system maximum is to carry out photograph taking to object of suspicion, and because photographed data amount is larger, remote transmission speed is slow, therefore can not realize magnanimity identification, and, be only can only candid photos to compare in the suspection stage.

Therefore, there is defect in prior art, needs to improve.

Utility model content

Technical problem to be solved in the utility model is: providing a kind of can magnanimity find and identification Given Face image, and data rate is fast, and the high magnanimity people's face of recognition accuracy is found and recognition system.

The technical solution of the utility model is as follows: a kind of magnanimity people's face is found and recognition system, comprise interconnective fore device and background process device, described fore device comprises continuous videos stream video camera and the candid camera being connected with recognition unit, and described recognition unit arranges recognition of face and cutting module and first network transport module; Described background process device comprises front end processor, operator's console and the memory storage being connected with server, and, the second network transport module that described front end processor setting is corresponding with described first network delivery module.

Be applied to technique scheme, in described magnanimity people's face discovery and recognition system, described candid camera is set to high definition candid cameras more than 5,000,000 pixels.

Be applied to each technique scheme, in described magnanimity people's face discovery and recognition system, in described continuous videos stream video camera and described candid camera, be respectively arranged with SD storage card.

Be applied to each technique scheme, in described magnanimity people's face discovery and recognition system, described wireless transmitting module arranges respectively the wireless delivery module of 3G or 4G or WiFi.

Adopt such scheme, the utility model is by arranging recognition unit, by recognition unit, the scene picture of capturing is identified, in confirming scene picture, include after people's face, again in the scene picture of capturing by the picture cutting of people face part out, and then be sent to long-range server and contrast identification, so, wirelessly transmitting data amount reduces, transmission speed is very fast, can realize the identification of people's face magnanimity, and, after server carries out the automatic contrast of database picture, can also again carry out manpower comparing pair, thereby realize 100% comparison reliability, recognition accuracy is high.

Accompanying drawing explanation

Fig. 1 is connection diagram of the present utility model.

Embodiment

Below in conjunction with the drawings and specific embodiments, the utility model is elaborated.

The present embodiment provides a kind of magnanimity people's face to find and recognition system, be applied to find offender, can be in the crowd of magnanimity, find and identification Given Face image information, thereby by the geographic position of taking, particular persons is found out, in the boundless and indistinct sea of faces, found those hiding offenders, there is the technique effect that recognition efficiency is high, deal with data amount is large.

As shown in Figure 1, magnanimity people's face is found to comprise interconnective fore device and background process device with recognition system, wherein, fore device comprises recognition unit, continuous videos stream video camera and candid camera, wherein, continuous videos stream video camera and candid camera interconnect, and, be connected with recognition unit respectively, continuous videos stream video camera is for photographed scene video recording, when there is object of which movement in scene video recording, continuous videos stream video camera sends instruction to candid camera, candid camera is captured scene picture, and scene picture is sent to recognition unit, and in order to ensure the sharpness of capturing, candid camera adopts high-definition cameras more than 5,000,000 pixels.

And, recognition unit arranges recognition of face and cutting module and first network transport module, in recognition of face and cutting module, be provided with integrated circuit (IC) chip, the scene picture that integrated circuit (IC) chip realizes capturing by the computer program of its storage is identified and is cut, wherein, recognition of face and cutting module are when identification scene picture, mainly whether scene picture to be comprised to people's face identify, while including people's face in identification scene picture, by the picture cutting of people face part out, and, first network transport module comprises wireless delivery module and cable network module, usually, be set to wireless delivery module, for example, 3G, or 4G, or the wireless delivery module of WiFi, cutting people face part picture is out sent in long-range server by first network transport module, when the unmanned face of identification, do not carry out cutting, when whether identification has people's face, can specifically identify by people's feature, for example, people's hair, eyes, ear, the imaging features of nose etc. is specifically identified.

And, background process device comprises server, front end processor, operator's console and memory storage, front end processor, operator's console and memory storage are connected with server respectively, wherein, server is a PC main frame, and operator's console includes display, mouse and keyboard, can carry out operations server by operator's console, by keyboard and mouse input instruction, thereby pass through display displayed scene, record a video and scene picture.

And, memory storage can be independent memory device, also can be the storage unit that is fixed on server inside, front end processor arranges the second network transport module corresponding with described first network delivery module, front end processor is mainly used in receiving wired or wireless data and sends server to, and the data that server is sent are launched by wired or wireless mode, for example, can be set to a transmitting set module, second network transport module comprises wireless delivery module and cable network module, usually, be set to wireless delivery module, for example, 3G, or 4G, or the wireless delivery module of WiFi.

So, first network delivery module is sent to the people face part picture after cutting in second network transport module, second network delivery module is sent to people's face picture in server again, and, after people's face picture being judged by server, people's face picture is judged as to a bad actor, can send instruction to recognition unit by second network delivery module and first network delivery module, the scene that continuous videos is flowed to video camera shooting by recognition unit is recorded a video, and the scene picture captured of candid camera sends the server preservation of filing together to, to offer law enfrocement official, use.

These are only preferred embodiment of the present utility model, be not limited to the utility model, all any modifications of doing within spirit of the present utility model and principle, be equal to and replace and improvement etc., within all should being included in protection domain of the present utility model.

Claims

1. magnanimity people face is found and a recognition system, comprises interconnective fore device and background process device, it is characterized in that:

Described fore device comprises continuous videos stream video camera and the candid camera being connected with recognition unit, and described recognition unit arranges recognition of face and cutting module and first network transport module;

Described background process device comprises front end processor, operator's console and the memory storage being connected with server, and, the second network transport module that described front end processor setting is corresponding with described first network delivery module.

2. magnanimity people face is found and recognition system according to claim 1, it is characterized in that: described candid camera is set to high definition candid cameras more than 5,000,000 pixels.

3. magnanimity people face is found and recognition system according to claim 2, it is characterized in that: in described continuous videos stream video camera and described candid camera, be respectively arranged with SD storage card.

4. according to the arbitrary described magnanimity people's face of claim 1-3, find and recognition system, it is characterized in that: described first network transport module and second network transport module arrange respectively the wireless delivery module of 3G or 4G or WiFi.