WO2010031213A1 - 一种智能身份识别方法及其实现该方法的识别终端 - Google Patents
一种智能身份识别方法及其实现该方法的识别终端 Download PDFInfo
- Publication number
- WO2010031213A1 WO2010031213A1 PCT/CN2008/072367 CN2008072367W WO2010031213A1 WO 2010031213 A1 WO2010031213 A1 WO 2010031213A1 CN 2008072367 W CN2008072367 W CN 2008072367W WO 2010031213 A1 WO2010031213 A1 WO 2010031213A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- face
- central processing
- processing unit
- face image
- interface
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
Definitions
- the invention discloses a smart identification technology, which is particularly suitable for a bank ATM machine and a bank counter.
- Intelligent identification method for security inspection posts that need to quickly identify the identity of the object to be detected, such as customs border inspection, airport security inspection, security confidentiality department, entry control system of each relevant unit, and identification terminal for realizing the method, belonging to public safety intelligence Early warning technology.
- the usual technical means used in urban security work is to arrange cameras in the city, establish monitoring centers, and provide video equipment.
- the shortcoming of this technical solution is post-processing, that is, after the case occurs, by looking at the video, looking for images of criminal suspects, providing clues for the public security department to solve the case.
- the result of using such a technical solution is that the case has already occurred, the loss has already been caused, the suspect has escaped, the case cannot be solved, and the cost of solving the case is increased; and the city security monitoring network needs to arrange a large number of cameras, monitor 24 hours, and manually monitor , or video mode implementation.
- the present invention provides a smart identification.
- Method and an identification terminal for implementing the same, using an existing camera, or in the present invention The camera is installed as a face information collecting device, and the face information is collected, and the face information collected by the camera is input into the central processing unit, and the face image information is pre-stored in the central processing unit, and is collected in the central processing unit.
- the face information is compared with the pre-stored face image information. When the comparison result exceeds the confidence level, the central processing unit drives the alarm device to alert, or informs the auditor to arrest the person or further confirm it according to the auxiliary decision information. .
- the similarity between the input face image and the registered face image feature is calculated, and then a fixed threshold is set, and if the maximum similarity is greater than the threshold, the recognition is performed. , rejection recognition less than the threshold.
- the method based on fixed threshold has obvious shortcomings. Because there are various changes in face images, the individual fixed thresholds are very difficult to determine. Too low will result in a high error rate. Too high will lead to higher rejection. rate.
- the present invention proposes a new confidence calculation method, which is not based on a fixed matching degree, but is dynamically determined by the test face image, and can well solve the threshold problem of confidence.
- a smart identification method includes:
- A provided with an image capturing device for taking a portrait or having an interface for connecting with an existing camera device and transmitting data;
- [7] B There is a central processing unit, and the central processing unit stores the face image information, and is provided with a confidence level indicating the degree of similarity between the face image to be recognized and the stored face image, and the camera device is implemented. Collecting face information, performing face recognition, and comparing with the face image information preset in the central processing unit. When the comparison result exceeds the confidence level, the central processing unit drives the alarm device to alarm, or will be identified. The face image is transmitted to the auditor and confirmed according to the auxiliary information;
- [8] C There is a transmission unit connected to the central processing unit.
- the central processing unit cannot judge the recognized face ⁇ through the built-in face image information, and transmits the information to the remote database server through the transmission unit for judgment.
- An identification terminal for implementing a smart identification method comprising:
- A camera unit or camera interface: used to collect face information and transmit it to the central processing unit, which is the camera or the interface connected with the existing camera;
- [11] B It is used to identify face information and compare it with the face image information stored in the memory. And a central processing unit that determines the confidence of the recognized face;
- [13] D a data transmission unit connected to the central processing unit for connecting with the remote database server for further determining the face information that cannot be judged by the central processing unit, and updating the face image information stored in the storage unit ;
- the transmission unit and the remote database can perform data transmission by means of wired communication or wireless communication.
- the central processing unit performs face recognition, uses human eye positioning, and performs face feature recognition in combination with the local texture and overall appearance of the face.
- the human eye positioning is to establish a model for the texture of a large number of human eye data through statistical learning methods, and then perform human eye localization according to the matching degree between the identified image region and the model, and establish the blur of the human eye position.
- the membership function constrains the positioning result.
- the local texture and overall appearance of the face are performed for face feature recognition.
- the face is analyzed by fuzzy pattern recognition and neural network theory. Firstly, a large number of face data are statistically analyzed to obtain face texture and face shape.
- the model uses the fuzzy similar matrix to construct the feature vector, extracts the facial features, and then uses the neural network to train the extracted features, and finally obtains the face recognition model.
- the human face positioning and feature extraction are performed on the input recognized face image, and the features of the face image information preset in the central processing unit are compared one by one to calculate a corresponding one.
- the series matching value based on one or more matching values, calculates the confidence.
- An analog-to-digital conversion unit is connected between the camera unit or the camera interface and the central processing unit.
- the data transmission unit is a USB interface, or an RS232 interface, or a wireless communication interface.
- the storage unit is RAM or FLASHROM or the same with RAM and FLASHROM.
- the beneficial effects of the present invention are as follows: First, through automatic intelligent screening technology, automatic searching of a large number of people is realized, the manual screening range is reduced, and the work intensity of relevant inspectors is greatly reduced; second, through automatic intelligent identification technology Auxiliary decision-making information greatly improves the accuracy of screening, and changes the defect that the single biometric automatic recognition technology cannot be applied because the recognition accuracy is not high. Third, the recognition accuracy is adjustable. . Through the adjustment of the automatic recognition accuracy, the recognition speed can be accelerated (due to the auxiliary decision-making process, the automatic recognition accuracy can be appropriately reduced); Fourth, the practicality is strong. The identification accuracy and recognition speed of the current single biometric intelligent recognition technology are still a certain distance from the actual application.
- the invention can better compensate for the defects of the current technology, thereby realizing practicalization and productization;
- Fifth the terminal device It is easy to use, can be used as a video input by the camera configured by the camera, or by using the video signal collected by the installed camera system. It can be used locally for banks, customs, border inspection and other units, and can be networked to form a wide-area monitoring system.
- FIG. 1 is a block diagram of a system architecture of the present invention.
- FIG. 2 is a flow chart of the working process of the present invention.
- FIG. 3 is a schematic diagram of a circuit of a video input portion according to an embodiment of the present invention.
- FIG. 4 is a schematic diagram of a circuit of a voice input/output section according to an embodiment of the present invention.
- FIG. 5 is a circuit schematic diagram of a portion of a programmable logic device in accordance with an embodiment of the present invention.
- FIG. 6 is a schematic diagram of a circuit of a video signal output portion according to an embodiment of the present invention.
- FIG. 7 is a schematic diagram of a circuit of a network connection portion according to an embodiment of the present invention.
- FIG. 8 is a schematic circuit diagram of a portion of a clock of the embodiment of the present invention.
- FIG. 9 is a schematic diagram of a transmission interface circuit according to an embodiment of the present invention.
- FIG. 10 is a schematic diagram of a circuit portion of a FLASH ROM according to an embodiment of the present invention.
- FIG. 11 is a schematic diagram of a circuit of a memory portion according to an embodiment of the present invention.
- FIG. 12 is a schematic circuit diagram of a portion of a hard disk interface according to an embodiment of the present invention.
- FIG. 13 is a circuit schematic diagram of a power management unit according to an embodiment of the present invention.
- FIG. 14 is a schematic circuit diagram of a power management unit 2 according to an embodiment of the present invention.
- FIG. 15 is a circuit diagram of a first part of a central processing unit according to Embodiment 2 of the present invention.
- FIG. 16 is a circuit diagram of a second part of a central processing unit according to Embodiment 2 of the present invention.
- FIG. 17 is a schematic diagram of a circuit of a video input portion according to Embodiment 2 of the present invention.
- FIG. 18 is a schematic diagram of a circuit of a video output portion according to Embodiment 2 of the present invention.
- FIG. 19 is a schematic diagram of a voice input/output circuit according to a second embodiment of the present invention.
- FIG. 20 is a schematic diagram of a display interface circuit according to a second embodiment of the present invention.
- FIG. 21 is a schematic diagram of a USB interface circuit according to Embodiment 2 of the present invention.
- FIG. 22 is a schematic circuit diagram of a circuit portion of an SD card according to a second embodiment of the present invention.
- FIG. 23 is a schematic circuit diagram of a wired network connection part according to Embodiment 2 of the present invention.
- FIG. 24 is a schematic circuit diagram of a wireless network connection part according to Embodiment 2 of the present invention.
- FIG. 25 is a schematic circuit diagram of a power management unit according to Embodiment 2 of the present invention.
- the present embodiment is a preferred embodiment of the present invention, and other principles and basic structures are the same or similar to those of the present embodiment, and are all within the scope of the present invention.
- the present invention includes a central processing unit, a storage unit, a data transmission unit, a camera interface, and a power management unit.
- the face information collected by the camera interface is input.
- the digital conversion unit converts the digital signal into a digital signal and inputs it to the central processing unit for processing.
- the central processing unit is connected with a storage unit, the storage unit stores the face image information, and the central processing unit compares the face information and the storage transmitted by the camera interface.
- the face image information stored therein is connected to the central processing unit with a data transmission unit. When the face image information stored in the storage unit is insufficient to determine the recognized face, the recognized face information is transmitted to the data transmission unit.
- the remote database server makes further judgments, and the power management unit supplies power to the various units in the present invention.
- the present invention collects face information through a camera mounted on a bank ATM machine or other surveillance camera, and collects the face through the camera interface.
- the information is transmitted to the central processing unit, or the imaging device is mounted on the terminal device of the present invention, the facial information is collected by the imaging device, and the collected facial information is transmitted to the central processing unit, and the central processing unit uses the human eye.
- the texture of the large amount of human eye data is modeled by the statistical learning method, and then the human eye positioning is performed according to the matching degree between the search area and the model.
- Establishing a fuzzy membership function of the human eye position constrains the search results, resulting in very stable positioning results.
- the central processing unit locates the face region and the contour according to the position of the human eye, and recognizes and extracts the overall appearance and the local texture feature of the human face.
- the fuzzy pattern recognition and the neural network theory are comprehensively used to analyze the face.
- the neural network focuses on the micro-network structure of the human brain.
- Parallel analysis methods formed by learning, self-organization and nonlinear dynamics theory can process schema information that cannot be linguistically, while fuzzy set theory focuses on available languages and As a representative macro function of the human brain, according to the introduced membership function, the information containing fuzzy logic is processed.
- a large amount of face data is statistically analyzed to obtain a face texture, a face model, and a fuzzy similarity.
- the matrix constructs the feature vector, extracts the facial features, and then uses the neural network to train the extracted features, and finally obtains the model of face recognition.
- the central processing unit compares the recognized face information with the face image information of the storage unit, and determines the confidence of the detected face (the confidence is the recognized face information and the person stored in the storage unit)
- the face image information is compared in terms of local texture and overall appearance, thereby obtaining a value indicating the degree of approximation between the recognized face information and the stored face image information, the confidence range is 0 ⁇ 1, preset
- the value of the confidence value in the central processing unit is adjustable, and a method for calculating the confidence is as follows: comparing the feature of the face image to be recognized with the feature of the face image stored in the memory to obtain a matching value greater than zero.
- the central processing unit drives the alarm device to alarm, or the drive management system informs the auditor
- the identified person performs further confirmation or arrest, etc.
- 2 000-5000 commonly used or currently being captured face information can be stored in the storage device, and stored in the storage device of the present invention.
- the face image information is insufficient to judge the recognized person, and the face information of the recognized person can be transmitted to the remote database server through the communication interface for further identification and confirmation, and the face stored in the memory can also be communicated through the communication interface.
- the image information is automatically updated.
- a specific implementation circuit structure of the present invention will be specifically described below in conjunction with a specific circuit diagram.
- Two specific embodiments are given in the present invention, one is a system focusing on comparison operation and real processing, and mainly uses a DSP system as a central processing unit; the other uses a powerful Z228 chip as a central processing.
- the unit focusing on image compression and processing, will be described in detail below for each of the two embodiments.
- Embodiment 1 the DSP system is used as the central processing unit main chip, and referring to FIG. 3 to FIG. 13, in this embodiment, the DSP chip U22 uses the audio and video digital signal processing function.
- the specific model of the chip is TMS320DM6446.
- the specific implementation can select the specific type of DSP chip according to the specific situation and the occasion.
- the DSP chip U22 is divided into several independents for the convenience of description. The description of the vertical part is actually a chip.
- the video input circuit portion of the present invention is shown in FIG. 3.
- the video signal ie, the face information collected by the camera or the camera
- the interface J14 and J27 is input to the video decoder U35 and the video decoder U34, respectively, and the video signals collected by the video decoder U35 and the video decoder U34 are decoded, and then input to the DSP chip U22, the video decoder U35 and the video decoding respectively.
- the function pins (such as the ⁇ clock pin, control pin, etc.) in the U34 are synthesized and connected to the corresponding pins of the DSP chip U22.
- the specifics of the video decoder U35 and the video decoder U34 are given.
- the signal is TVP5150PBS.
- the voice input/output part circuit of the present invention is shown in FIG. 4.
- audio information is input through the input audio interface J18, output through the output audio interface J16, and input audio.
- the interface J18 and the output audio interface J16 are respectively connected to the audio signal processing chip U23.
- the audio signal processing chip U23 uses the specific model TLV320AIC23BPW, and can be replaced by other types of audio signal processing chips.
- the input/output interface of the audio signal processing chip U23 is connected to the DSP chip U22 via the latch U3.
- FIG. 5 is a circuit part of a programmable logic device according to the present invention.
- a programmable circuit device uses a CPLD chip U7, and a specific model selected by the CPLD chip U7 is XC2C128. After implementation, you can also use other types of CPLD chips instead.
- FIG. 6 is a circuit diagram of a video signal output portion of the present invention.
- a video signal is driven by a video interface to display the display.
- the analog video output interface of the DSP chip U22 inputs four signals respectively, and the four signals are respectively amplified by the operational amplifier circuit.
- the operational amplifier circuit is operated by an operational amplifier with a switching function and a peripheral capacitor resistor. The output of the operational amplifier circuit is connected to a display interface for driving the display.
- FIG. 7 is a network connection part of the present invention.
- the RJ45 crystal head interface T1 is connected to an external network, and the serial data transceiver end of the crystal head interface T1 is connected to 10/.
- the serial data transceiver terminal of the 100M network transceiver U52, each data line is connected to the +2.5V power supply through a pull-up resistor, and the data output terminal, the clock terminal and the control signal terminal of the 10/100M network transceiver U52 are respectively The corresponding interface of the DSP chip U 22 is connected.
- the specific model of the 10/100M network transceiver U52 is BC M5221, and the specific implementation may be replaced by other models with the same function. Referring to FIG.
- FIG. 8 is a crotch clock part of the present invention.
- a general-purpose chopper generator U58 is used as a cuckoo clock chip.
- a general-purpose chopper generator is used.
- the U58 uses a chip of the type CY22381.
- the general-purpose chirp generator U58 generates a chirped pulse signal with a frequency of 14.31818M, 25M and 3.57M to provide a chirp signal to the system.
- FIG. 9 is an interface circuit diagram of the present invention.
- various interface forms such as USB, I2C, R S485, and RS232 are provided and externally connected.
- the RS2 32 interface circuit uses the RS232 driver transceiver U13 of the type MAX3221CPWR, the RS232 interface J3 and the serial interface of the RS232 driver transceiver U13, and the output port of the RS232 driver transceiver U13.
- the RS485 interface circuit uses the R S485 driver transceiver U15 of the M41T11M6, the RS485 interface is connected to the serial interface of the RS485 driver transceiver U15, and the RS4 85 drives the output of the transceiver U15.
- the port is connected to the RS485 input and output end of the DSP chip U22.
- the I2C bus interface is implemented by the I2C bus encoder of the model PCA9306DCTT; the USB interface is directly connected to the USB interface end of the DSP chip U22.
- FIG. 10 is provided with a FLASHROM chip U28, and the FLASHROM chip U28 is connected by a dual bus three-state lock.
- the memory U29 is connected to the data terminal of the DSP chip U22, and the data terminal of the DSP chip U22 is also connected to the universal input/output port through the dual bus tristate latch U68.
- the dual bus tristate latch U29 and The dual-bus tristate latch U68 is modeled after the SN74AVCB16245VR. For other implementations, other types of tristate latches can be used.
- the data bus and control bus of the memory chip U1 and the memory chip U2 are respectively connected to the data end of the DSP chip U22 and the corresponding data output end.
- the data bus of the AT A hard disk interface is connected to the data end of the DSP chip U22 through the tristate latch U42, and the control bus of the ATA hard disk interface is corresponding to the DSP chip U22 through the dual bus tristate latch U27.
- the control signal output terminal is connected, and the data end and the control end of the SD card interface J8 are respectively connected with the corresponding ports of the DSP chip U22.
- the power management unit of the present invention is shown in FIG. 13 and FIG. 14.
- the +5V input interface S1 is used as the power input
- the special power management chip U10 of the FPGA and the DSP is used to supply power to the system.
- the power management circuit is used to detect the output voltage of the power management chip U10.
- the power management chip U10 is TPS75003RHLR
- the power management circuit is TPS3808G09DBVR°
- Embodiment 2 Referring to FIG. 14 to FIG. 24, a processing chip of the ARM core is taken as an example for specific description. In this embodiment, the processing chip uses a chip of model Z228. In the drawing, the processing chip is divided into two parts, which are respectively shown in Fig. 14 and Fig. 15.
- FIG. 16 is a circuit diagram of a video input portion of the embodiment.
- a video signal is input to a video decoder U2 through a video input terminal J1 for decoding, and the data of the video decoder U2 is used.
- the output end and the control end are respectively connected to the corresponding interfaces of the processing chip.
- the model of the video decoder U 2 is implemented by using the TVP5150A-TQFP32, and may be replaced by other types of video decoder chips.
- FIG. 17 is a circuit diagram of a video output portion of the embodiment.
- a data input end of the video encoder U3 is connected to a data interface corresponding to the processing chip, and a video signal of the video encoder U3 is connected.
- the output of the video output terminal J3 is used.
- the model of the video encoder U32 is implemented by using the SAA7120 H. Alternatively, other types of video encoder chips may be used instead.
- FIG. 18 is a circuit diagram of a voice input/output circuit in the embodiment.
- an audio codec U4 with a touch screen coding function is used for audio signal processing, and an audio codec U4 data is used.
- the audio signal input end of the audio codec U4 is connected to the microphone, and the audio signal output end of the audio codec U4 is connected to the speaker.
- the audio is used.
- the codec U4 model is WM9712L, which can be replaced by other models of audio decoder chips.
- FIG. 19 shows the LCD interface J13 in the embodiment, and the LCD interface J13 is an LCD interface with a touch screen, and the LCD data terminal and the control terminal are respectively connected with corresponding interfaces of the processing chip, LCD The touch screen data terminals of the interface J13 are respectively connected to the corresponding interfaces of the audio codec U4 of FIG.
- FIG. 20 is a circuit diagram of a portion of a USB interface circuit of the present embodiment.
- a USB transceiver U13 is used as a USB control chip, and a data input terminal of the USB transceiver U13 and a USB interface are used. The data terminal is connected, and the data output end and the control end of the USB transceiver U13 are respectively connected to the corresponding interfaces of the processing chip.
- the USB transceiver U13 model used is ISP1301, and the specific implementation may also use other models.
- the USB control chip is replaced.
- FIG. 21 is a partial circuit diagram of the SD card circuit of the embodiment.
- SD The card is connected to the system through the SD card interface CN6, and the data end of the SD card interface CN6 is connected to the corresponding interface of the processing chip.
- the invention also has a wired network connection and a wireless network connection.
- the wired network connection is implemented by the Ethernet controller U1, and the data bus, the address bus, and the control port of the Ethernet controller U1 are respectively connected to the corresponding interfaces of the processing chip.
- the Ethernet controller U1 model used is LAN9115.
- Ethernet controller chips For specific implementation, other types of Ethernet controller chips can be used instead.
- the wireless network connection is implemented by WIFI.
- the data end and the control end of the WIF I controller chip are respectively connected to corresponding interfaces of the processing chip.
- a 12V power supply is used as a power input, and a 12V power supply is converted into a +5V power supply through a regulator (in this embodiment, the model is LM2679S-5), and the power is supplied through the 3.3V.
- the power regulator (in this embodiment, the model is XC6210A-33) converts the 5V power supply to the +3.3V power supply to provide 3.3V power to the system, and uses the power management chip model RT9907-QFN24 as the main power supply of the system. Management chip.
- the invention can be widely applied to the fields of bank, public security, customs, exit and entry management, entrance threshold system of various relevant units, etc. Through the invention, the hidden dangers can be quickly discovered, and illegal criminal acts and suspects can be found, and the public security department can be assisted in solving the crime. .
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Description
说明书 一种智能身份识别方法及其实现该方法的识别终端 技术领域
[1] 本发明公开一种智能身份识别技术, 是一种特别适用于银行 ATM机、 银行柜台
、 海关边检、 机场安检,安全机要部门,各有关单位入口门禁系统等需要快速甄别 被检测对象身份的安全检査岗位的智能身份识别方法及其实现该方法的识别终 端, 属于公共安全智能预警技术领域。
背景技术
[2] 随着构建和谐社会、 平安城市工作开展, 城市安保问题被提到一定高度, 目前
, 城市安保工作中釆用的通常技术手段是在城市布置摄像头, 建立监控中心, 配备录象设备。 这种技术方案的缺陷是事后处理, 即当案件发生后, 通过调看 录象, 寻找犯罪嫌疑人影象, 为公安部门破案提供线索。 运用这样技术方案的 结果是案件已经发生, 损失已经造成, 犯罪嫌疑人已经逃跑, 不能及吋破案, 破案成本提高; 而且城市安全监控网需要布置大量的摄像头, 进行 24小吋监控 , 通过人工监视, 或录象方式实施。 如果釆用人工监视吋, 需大量的安保人员 2 4小吋守侯, 并且一个人也只能监视有限的画面; 录象则会产生了大量的视频信 息 (录象) , 信息具有吋效性, 如得不到及吋处理, 必将产生大量的信息垃圾 , 因此, 釆用现行的安保技术方案, 效率极低, 难以胜任构建平安城市的重任 。 随着科学技术的不断进步, 伪造证件、 整形化装技术等也在不断进步, 犯罪 嫌疑人可通过伪造证件, 或整形或化装, 改变形象逃避检査, 这样给检査或监 视人员的工作带来了很大困难。 综上所述, 可以看出, 现行的城市公共安全安 保手段, 已难以适应新形势下安保工作的要求, 急需一种新技术手段, 来改变 现行技术方案的缺陷, 以满足城市安全保卫工作的需要。
发明内容
[3] 针对上述提到的现有技术中的城市安保问题中需要釆用大量摄像头, 且为事后 处理, 难以实现事前预防, 吋效性差、 效率低等缺点, 本发明提供一种智能身 份识别方法及其实现该方法的识别终端, 其釆用现有的摄像头, 或在本发明中
安装摄像头作为人脸信息釆集装置, 进行人脸信息釆集, 将釆集到的人脸信息 输入到中央处理单元, 中央处理单元内预存有人脸图像信息, 在中央处理单元 内将釆集到的人脸信息与预存的人脸图像信息进行比较, 当比较结果超过置信 度吋, 中央处理单元驱动报警装置报警, 或通知稽査人员对此人进行缉捕或根 据辅助决策信息对其进行进一步确认。
[4] 在现有的人脸识别方法中, 都是计算输入人脸图像和已注册人脸图像特征之间 的相似度, 然后设定一个固定的阈值, 若最大相似度大于阈值的进行识别, 小 于阈值的拒绝识别。 基于固定阈值的方法有明显的缺点, 因为人脸图像有各种 各样的变化, 因此单独的固定阈值非常难确定, 过低将导致很高的错误率, 过 高将导致较高的拒识率。 本发明提出了一种新的置信度计算方法, 该置信度不 是基于固定的匹配度, 而是动态的由测试人脸图像决定, 可以很好的解决置信 度的阈值问题。
[5] 本发明解决其技术问题釆用的技术方案是: 一种智能身份识别方法, 该方法包 括:
[6] A、 设有一用于摄取人像的摄像装置或设有与现有摄像装置连接、 传输数据的 接口;
[7] B、 设有一中央处理单元, 中央处理单元内存储有人脸图像信息, 并设置有表 征待识别的人脸图像与存储的人脸图像之间的相似程度的置信度, 摄像装置实 吋釆集人脸信息, 进行人脸识别, 并与预置在中央处理单元内的人脸图像信息 进行实吋比较, 当比较结果超过置信度吋, 中央处理单元驱动报警装置报警, 或将被识别人脸图像传输给稽査人员, 根据辅助信息加以确认;
[8] C、 与中央处理单元连接有传输单元, 中央处理单元通过内置的人脸图像信息 不能判断识别的人脸吋, 通过传输单元将信息传输至远程数据库服务器进行判 断。
[9] 一种实现智能身份识别方法的识别终端, 终端包括:
[10] A、 摄像单元或摄像接口: 用于釆集人脸信息并传输给中央处理单元, 为摄像 头或与现有摄像头连接的接口;
[11] B、 用于识别人脸信息, 并与存储在存储器内的人脸图像信息进行实吋比较,
并实吋判断被识别人脸的置信度的中央处理单元;
[12] C、 存储有人脸图像信息的存储单元;
[13] D、 与中央处理单元连接的, 用于与远程数据库服务器连接以便于进一步判断 不能通过中央处理单元判断的人脸信息, 并可更新存储单元内存储的人脸图像 信息的数据传输单元;
[14] E、 给系统中各个单元供电的电源管理单元。
[15] 本发明解决其技术问题釆用的技术方案进一步还包括:
[16] 所述的传输单元与远程数据库可釆用有线通信或无线通信的方式进行数据传输
[17] 所述的中央处理单元进行人脸识别吋, 釆用人眼定位, 并结合人脸的局部纹理 和整体外观进行人脸特征识别。
[18] 所述的人眼定位是通过统计学习方法, 对大量人眼数据的纹理进行建立模型, 然后根据被识别图像区域与模型的匹配度进行人眼定位, 同吋建立人眼位置的 模糊隶属度函数对定位结果进行约束。
[19] 所述的人脸的局部纹理和整体外观进行人脸特征识别通过模糊模式识别和神经 网络理论对人脸进行分析, 先对大量人脸数据进行统计分析, 得到人脸纹理, 脸型的模型, 利用模糊相似矩阵构造特征向量, 进行人脸特征的提取, 然后再 对提取好的特征利用神经网络进行训练, 最终得到人脸识别的模型。
[20] 在进行人脸识别吋, 对输入的识别人脸图像进行人眼定位以及特征提取, 并与 预置在中央处理单元内的人脸图像信息的特征逐一进行对比, 计算出对应的一 系列匹配值, 根据一个以上或者全部匹配值, 计算出置信度。
[21] 所述的摄像单元或摄像接口与中央处理单元之间连接有模数转换单元。
[22] 所述的数据传输单元为 USB接口, 或 RS232接口, 或无线通信接口。
[23] 所述的存储单元为 RAM或 FLASHROM或同吋设有 RAM和 FLASHROM。
[24] 本发明的有益效果是: 第一, 通过自动智能甄别技术, 实现海量人群的自动搜 索, 缩小人工甄别范围, 大大减轻相关检査人员的工作强度; 第二, 通过自动 智能识别技术加辅助决策信息, 大大提高了甄别的准确性, 改变了单一生物特 征自动识别技术由于识别精度不高, 而不能应用的缺陷; 第三, 识别精度可调
。 通过自动识别精度的调整, 可加快识别速度 (由于有辅助决策过程, 所以可 适当降低自动识别精度) ; 第四, 实用性强。 现行的单一生物特征智能识别技 术的识别精度和识别速度离实际应用还有一定的距离, 利用本发明可较好地弥 补现行技术的缺陷, 从而实现实用化、 产品化; 第五, 本终端装置使用方便, 可通过本机配置的摄像头, 或利用已安装的摄像系统釆集的视频信号作为视频 输入, 即可为银行、 海关、 边检等单位局部使用, 又可联网形成广域监控体系
。 安装方便灵活。
附图说明
[25] 图 1为本发明系统架构框图。
[26] 图 2为本发明工作过程流程图。
[27] 图 3为本发明实施例-视频输入部分电路原理图。
[28] 图 4为本发明实施例-语音输入 /输出部分电路原理图。
[29] 图 5为本发明实施例-可编程逻辑器件部分的电路原理图。
[30] 图 6为本发明实施例-视频信号输出部分电路原理图。
[31] 图 7为本发明实施例- 网络连接部分电路原理图。
[32] 图 8为本发明实施例-吋钟部分电路原理图。
[33] 图 9为本发明实施例-传输接口电路原理图。
[34] 图 10为本发明实施例 —FLASH ROM部分电路原理图。
[35] 图 11为本发明实施例一内存部分电路原理图。
[36] 图 12为本发明实施例一硬盘接口部分电路原理图。
[37] 图 13为本发明实施例一电源管理单元一电路原理图。
[38] 图 14为本发明实施例一电源管理单元二电路原理图。
[39] 图 15为本发明实施例二中央处理单元第一部分电路图。
[40] 图 16为本发明实施例二中央处理单元第二部分电路图。
[41] 图 17为本发明实施例二视频输入部分电路原理图。
[42] 图 18为本发明实施例二视频输出部分电路原理图。
[43] 图 19为本发明实施例二语音输入输出电路原理图。
[44] 图 20为本发明实施例二显示接口电路原理图。
[45] 图 21为本发明实施例二 USB接口电路原理图。
[46] 图 22为本发明实施例二 SD卡连接部分电路原理图。
[47] 图 23为本发明实施例二有线网络连接部分电路原理图。
[48] 图 24为本发明实施例二无线网络连接部分电路原理图。
[49] 图 25为本发明实施例二电源管理单元电路原理图。
具体实施方式
[50] 本实施例为本发明优选实施方式, 其他凡其原理和基本结构与本实施例相同或 近似的, 均在本发明保护范围之内。
[51] 请参看附图 1, 本发明包括有中央处理单元、 存储单元、 数据传输单元、 摄像 接口、 电源管理单元, 本实施例中, 通过摄像接口输入釆集到的人脸信息, 经 过模数转换单元转换成数字信号后输入给中央处理单元进行处理, 中央处理单 元上连接有存储单元, 存储单元内存储有人脸图像信息, 中央处理单元用于比 较摄像接口传输来的人脸信息与存储内存储的人脸图像信息, 中央处理单元上 连接有数据传输单元, 当存储单元内存储的人脸图像信息不足以判断被识别的 人脸吋, 通过数据传输单元将被识别人脸信息传输给远程数据库服务器进行进 一步判断, 电源管理单元给本发明中各个单元供电。
[52] 下面结合附图 2对实现本发明的方法作进一步详述, 本发明通过银行 ATM机上 安装的摄像头或其他的监控摄像头等釆集人脸信息, 通过摄像接口将釆集到的 人脸信息传输给中央处理单元, 或者在本发明的终端装置上安装有摄像装置, 通过摄像装置釆集人脸信息, 并将釆集到的人脸信息传输给中央处理单元, 中 央处理单元釆用人眼定位在摄像头釆集的人脸信息上定位出人眼位置, 本发明 中通过统计学习方法, 对大量人眼数据的纹理进行建模, 然后根据搜索区域与 模型的匹配度进行人眼定位, 同吋建立人眼位置的模糊隶属度函数对搜索结果 进行约束, 从而得到非常稳定的定位结果。 中央处理单元根据人眼位置定位出 人脸区域及轮廓, 并进行识别, 提取出人脸的整体外观和局部纹理特征, 本发 明中综合运用模糊模式识别和神经网络理论来对人脸进行分析, 神经网络着眼 于人脑的微观网络结构, 通过学习、 自组织化和非线性动力学理论形成的并行 分析方法, 可处理无法语言化的模式信息, 而模糊集理论则着眼于可用语言和
概念作为代表的人脑的宏观功能, 按照引入的隶属度函数, 处理包含有模糊逻 辑性的信息, 本发明中对大量人脸数据进行统计分析, 得到人脸纹理, 脸型的 模型, 利用模糊相似矩阵构造特征向量, 进行人脸特征的提取, 然后再对提取 好的特征利用神经网络进行训练, 最终得到人脸识别的模型。 中央处理单元将 识别出的人脸信息与存储内存储单元的人脸图像信息进行比较, 实吋判断检测 识别的人脸的置信度 (置信度为识别的人脸信息与存储在存储单元内人脸图像 信息在局部纹理和整体外观方面进行比较, 从而得出的一个表征被识别的人脸 信息与存储的人脸图像信息之间的近似程度的值, 置信度范围为 0~1, 预置在中 央处理单元内的置信度值大小可调, 置信度的一种计算方法如下: 将待识别的 人脸图像特征与存储器内存储的人脸图像特征进行对比, 得出一个大于零的匹 配值, 然后将这些匹配值进行排序, 从大到小为 ml, ml, m3, m4, ···, 置信 度的计算公式为 k (ml-m2) /ml或者 kml/m2或者 kml/ (m2+m3+m4) (其中 k为 一常数) , 当被检测人脸的置信度高于预置的置信度吋, 中央处理单元驱动报 警装置进行报警, 或者驱动管理系统等通知稽査人员对被识别人员进行进一步 确认或进行缉捕等。 由于本发明终端的存储能力有限, 可在其存储装置内存储 2 000-5000个常用的或当前正在缉捕的人脸信息, 当本发明存储装置内存储的人脸 图像信息不足以判断被识别的人吋, 可将被识别的人的人脸信息通过通信接口 传输至远程数据库服务器进行进一步识别确认, 也可以通过通信接口对存储在 存储器内的人脸图像信息进行实吋自动更新。
[53] 下面将结合具体电路图, 对本发明的一个具体实施电路结构进行具体说明。 本 发明中给出两种具体实施方式, 一种为侧重于比较运算、 实吋处理的系统, 主 要釆用 DSP系统作为中央处理单元; 另一种釆用图像处理功能强大的 Z228芯片作 为中央处理单元, 侧重于图像压缩及处理, 下面将针对两种实施方式分别进行 详细描述。
[54] 实施例一: 本实施例中, 以 DSP系统作为中央处理单元主芯片, 请参看附图 3 至附图 13, 本实施例中, DSP芯片 U22釆用带音视频数字信号处理功能的芯片, 给出的具体型号为 TMS320DM6446 , 具体实施吋可根据具体情况和使用场合选 用具体型号的 DSP芯片, 本实施例中为了叙述方便, 将 DSP芯片 U22分成几个独
立的部分进行描述, 实际为一个芯片。
[55] 请参看附图 3, 附图 3中示出的为本发明的视频输入电路部分, 本实施例中, 视 频信号 (即摄像装置或摄像头釆集到的人脸信息) 通过接口 J14和 J27分别输入到 视频解码器 U35和视频解码器 U34中, 通过视频解码器 U35和视频解码器 U34对釆 集到的视频信号进行解码后, 分别输入给 DSP芯片 U22, 视频解码器 U35和视频 解码器 U34中的功能引脚 (如吋钟引脚、 控制引脚等) 合成一路与 DSP芯片 U22 相应的引脚连接, 本实施例中, 给出的视频解码器 U35和视频解码器 U34的具体 信号为 TVP5150PBS。
[56] 请参看附图 4, 附图 4中示出的为本发明的语音输入 /输出部分电路, 本实施例 中, 音频信息通过输入音频接口 J18输入, 通过输出音频接口 J16输出, 输入音频 接口 J18和输出音频接口 J16分别连接至音频信号处理芯片 U23, 本实施例中, 音 频信号处理芯片 U23釆用具体型号为 TLV320AIC23BPW, 具体实施吋, 也可釆 用其他型号的音频信号处理芯片代替。 音频信号处理芯片 U23的输入输出接口通 过锁存器 U3与 DSP芯片 U22连接。
[57] 请参看附图 5, 附图 5为本发明中可编程逻辑器件的电路部分, 本实施例中, 可 编程逻辑器件釆用 CPLD芯片 U7, CPLD芯片 U7选用的具体型号为 XC2C128 , 具 体实施吋, 也可釆用其他型号的 CPLD芯片代替。
[58] 请参看附图 6, 附图 6为本发明中视频信号输出部分电路, 本实施例中, 视频信 号通过视频接口驱动显示器进行显示。 本实施例中, DSP芯片 U22的模拟视频输 出接口分别输入四路信号, 四路信号分别通过运算放大器电路进行信号放大, 本实施例中, 运算放大器电路由带开关功能的运算放大器配合外围电容电阻组 成, 运算放大器电路的输出接至显示器接口, 用来驱动显示器。
[59] 请参看附图 7, 附图 7中为本发明网络连接部分, 本实施例中, 通过 RJ45水晶头 接口 T1与外部网络连接, 水晶头接口 T1的串行数据收发端连接至 10/100M网络收 发器 U52的串行数据收发端, 每个数据线上通过一个上拉电阻连接至 +2.5V电源 , 10/100M网络收发器 U52的数据输出端、 吋钟端及控制信号端分别与 DSP芯片 U 22相应的接口连接, 本实施例中, 10/100M网络收发器 U52釆用的具体型号为 BC M5221 , 具体实施吋, 也可釆用同样功能的其他型号代替。
[60] 请参看附图 8, 附图 8为本发明中的吋钟部分, 本实施例中釆用通用型吋钟发生 器 U58作为吋钟芯片, 本实施例中, 通用型吋钟发生器 U58釆用型号为 CY22381 的芯片, 通用型吋钟发生器 U58产生频率为 14.31818M、 25M和 3.57M的吋钟脉冲 信号, 给系统提供吋钟信号。
[61] 请参看附图 9, 附图 9为本发明的接口电路图, 本实施例中提供了 USB、 I2C、 R S485、 RS232等多种接口形式与外部连接。 请具体参看附图, 本实施例中, RS2 32接口电路釆用型号为 MAX3221CPWR的 RS232驱动收发器 U13, RS232接口 J3 与 RS232驱动收发器 U13的串行接口连接, RS232驱动收发器 U13的输出端口与 D SP芯片 U22的 RS232输入输出端连接; RS485接口电路釆用型号为 M41T11M6的 R S485驱动收发器 U15, RS485接口与 RS485驱动收发器 U15的串行接口连接, RS4 85驱动收发器 U15的输出端口与 DSP芯片 U22的 RS485输入输出端连接; 本实施 例中, 通过型号为 PCA9306DCTT的 I2C总线编码器实现 I2C总线接口; USB接口 直接与 DSP芯片 U22的 USB接口端连接。
[62] 请参看附图 10、 附图 11、 附图 12, 本发明中设有存储装置, 请参看附图 10, 附 图 10内设有 FLASHROM芯片 U28, FLASHROM芯片 U28通过双总线三态锁存器 U29与 DSP芯片 U22的数据端连接, DSP芯片 U22的数据端上还通过双总线三态锁 存器 U68连接有通用输入输出端口, 本实施例中, 双总线三态锁存器 U29和双总 线三态锁存器 U68的型号为 SN74AVCB16245VR, 具体实施吋, 也可釆用其他型 号的三态锁存器。 请参看附图 11, 内存芯片 U1和内存芯片 U2的数据总线和控制 总线分别与 DSP芯片 U22的数据端和相应的数据输出端连接。 请参看附图 12, AT A硬盘接口的数据总线通过三态锁存器 U42与 DSP芯片 U22的数据端连接, ATA 硬盘接口的控制总线通过双总线三态锁存器 U27与 DSP芯片 U22相应的控制信号 输出端连接, SD卡接口 J8的数据端和控制端分别与 DSP芯片 U22相应的端口连接
[63] 本发明的电源管理单元请参看附图 13、 图 14, 本实施例中, 釆用 +5V输入接口 S1作为电源输入, 利用 FPGA、 DSP的专门的电源管理芯片 U10为系统提供电源 , 并设有电源管理电路, 用来实吋检测电源管理芯片 U10的输出电压, 本实施例 中, 釆用的电源管理芯片 U10的型号为 TPS75003RHLR, 电源管理电路的型号为
TPS3808G09DBVR°
[64] 实施例二: 请参看附图 14至附图 24, 釆用 ARM内核的处理芯片为例进行具体 说明, 本实施例中, 处理芯片釆用型号为 Z228的芯片。 附图中, 将处理芯片分 成两个部分分别记载在附图 14和附图 15中。
[65] 请参看附图 16, 附图 16中为本实施例视频输入部分电路图, 本实施例中, 视频 信号通过视频输入端子 J1输入到视频解码器 U2中进行解码, 视频解码器 U2的数 据输出端与控制端分别与处理芯片相应的接口连接, 本实施例中, 视频解码器 U 2的型号釆用 TVP5150A-TQFP32, 具体实施吋, 也可釆用其他型号的视频解码器 芯片代替。
[66] 请参看附图 17, 附图 17为本实施例视频输出部分电路图, 本实施例中, 视频编 码器 U3的数据输入端与处理芯片相应的数据接口连接, 视频编码器 U3的视频信 号通过视频输出端子 J3输出, 本实施例中, 视频编码器 U32的型号釆用 SAA7120 H, 具体实施吋, 也可釆用其他型号的视频编码器芯片代替。
[67] 请参看附图 18, 附图 18为本实施例中语音输入输出电路图, 本实施例中, 釆用 带触摸屏编码功能的音频编解码器 U4进行音频信号处理, 音频编解码器 U4数据 端和控制端分别与处理芯片的相应接口连接, 音频编解码器 U4的音频信号输入 端与麦克连接, 音频编解码器 U4的音频信号输出端与扬声器连接, 本实施例中 , 釆用的音频编解码器 U4型号为 WM9712L, 具体实施吋, 也可釆用其他型号的 音频解码器芯片代替。
[68] 请参看附图 19, 附图 19中所示为本实施例中 LCD接口 J13, LCD接口 J13为带触 摸屏的 LCD接口, LCD数据端和控制端分别与处理芯片的相应接口连接, LCD 接口 J13的触摸屏数据端分别与图 18中音频编解码器 U4的相应接口连接。
[69] 请参看附图 20, 附图 20为本实施例的 USB接口电路部分电路图, 本实施例中釆 用 USB收发器 U13作为 USB控制芯片, USB收发器 U13的数据输入端与 USB接口 的数据端连接, USB收发器 U13的数据输出端和控制端分别与处理芯片的相应接 口连接, 本实施例中, 釆用的 USB收发器 U13型号为 ISP1301 , 具体实施吋, 也 可釆用其他型号的 USB控制芯片代替。
[70] 请参看附图 21, 附图 21为本实施例的 SD卡电路部分电路图, 本实施例中, SD
卡通过 SD卡接口 CN6与系统连接, SD卡接口 CN6的数据端与处理芯片的相应接 口连接。
本发明中还设有有线网络连接和无线网络连接。 请参看附图 22, 本实施例中, 有线网络连接通过以太网控制器 U1实现, 以太网控制器 U1的数据总线、 地址总 线和控制端口分别与处理芯片的相应接口连接, 本实施例中, 釆用的以太网控 制器 U1型号为 LAN9115,
具体实施吋, 也可釆用其他型号的以太网控制器芯片代替。
请参看附图 23, 本实施例中, 无线网络连接通过 WIFI实现, 本实施例中, WIF I控制器芯片的数据端和控制端分别与处理芯片的相应接口连接。
请参看附图 24, 本实施例中釆用 12V电源作为电源输入, 12V电源通过调整器 (本实施例中, 釆用的型号为 LM2679S-5) 转换成 +5V电源给系统供电, 通过 3.3 V电源调整器 (本实施例中, 釆用的型号为 XC6210A-33) 将 5V电源转换成 +3.3V 电源给系统提供 3.3V电源, 釆用型号为 RT9907-QFN24的电源管理芯片作为系统 的主电源管理芯片。
本发明可广泛应用于银行、 公安、 海关、 出入境管理、 各有关单位入口门襟系 统等领域, 通过本发明能快速的发现安全隐患, 及吋发现违法犯罪行为以及抓 获疑犯, 协助公安部门破案。
Claims
1、 一种智能身份识别方法, 其特征是: 所述的该方法包括:
A、 设有一用于摄取人像的摄像装置或设有与现有摄像装置连接、 传输数 据的接口;
B、 设有一中央处理单元, 中央处理单元内存储有人脸图像信息, 并设置 有表征待识别的人脸图像与存储的人脸图像之间的相似程度的置信度, 摄 像装置实吋釆集人脸信息, 进行人脸识别, 并与预置在中央处理单元内的 人脸图像信息进行实吋比较, 当比较结果超过置信度吋, 中央处理单元驱 动报警装置报警, 或将被识别人脸图像传输给稽査人员, 根据辅助信息加 以确认;
C、 与中央处理单元连接有传输单元, 中央处理单元通过内置的人脸图像 信息不能判断识别的人脸吋, 通过传输单元将信息传输至远程数据库服务 器进行判断。
2、 根据权利要求 1所述的智能身份识别方法, 其特征是: 所述的传输单元 与远程数据库可釆用有线通信或无线通信的方式进行数据传输。
3、 根据权利要求 1所述的智能身份识别方法, 其特征是: 所述的中央处理 单元进行人脸识别吋, 釆用人眼定位, 并结合人脸的局部纹理和整体外观 进行人脸特征识别。
4、 根据权利要求 3所述的智能身份识别方法, 其特征是: 所述的人眼定位 是通过统计学习方法, 对大量人眼数据的纹理进行建立模型, 然后根据被 识别图像区域与模型的匹配度进行人眼定位, 同吋建立人眼位置的模糊隶 属度函数对定位结果进行约束。
5、 根据权利要求 3所述的智能身份识别方法, 其特征是: 在进行人脸识别 吋, 对输入的识别人脸图像进行人眼定位以及特征提取, 并与预置在中央 处理单元内的人脸图像信息的特征逐一进行对比, 计算出对应的一系列匹 配值, 根据一个以上或者全部匹配值, 计算出置信度。
6、 一种实现如权利要求 1所述的智能身份识别方法的识别终端, 其特征是 : 所述的终端包括:
A、 摄像单元或摄像接口: 用于釆集人脸信息并传输给中央处理单元, 为 摄像头或与现有摄像头连接的接口;
B、 用于识别人脸信息, 并与存储在存储器内的人脸图像信息进行实吋比 较, 并实吋判断被识别人脸的置信度的中央处理单元;
C、 存储有人脸图像信息的存储单元;
D、 与中央处理单元连接的, 用于与远程数据库服务器连接以便于进一步 判断不能通过中央处理单元判断的人脸信息, 并可更新存储单元内存储的 人脸图像信息的数据传输单元;
E、 给系统中各个单元供电的电源管理单元。
7、 根据权利要求 7所述的识别终端, 其特征是: 所述的摄像单元或摄像接 口与中央处理单元之间连接有模数转换单元。
8、 根据权利要求 7所述的识别终端, 其特征是: 所述的数据传输单元为 US B接口, 或 RS232接口, 或无线通信接口。
9、 根据权利要求 7所述的识别终端, 其特征是: 所述的存储单元为 RAM或 FLASHROM或同吋设有 RAM和 FLASHROM。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2008/072367 WO2010031213A1 (zh) | 2008-09-16 | 2008-09-16 | 一种智能身份识别方法及其实现该方法的识别终端 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2008/072367 WO2010031213A1 (zh) | 2008-09-16 | 2008-09-16 | 一种智能身份识别方法及其实现该方法的识别终端 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010031213A1 true WO2010031213A1 (zh) | 2010-03-25 |
Family
ID=42039070
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2008/072367 WO2010031213A1 (zh) | 2008-09-16 | 2008-09-16 | 一种智能身份识别方法及其实现该方法的识别终端 |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2010031213A1 (zh) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106294246A (zh) * | 2016-09-30 | 2017-01-04 | 广州像素数据技术股份有限公司 | 一种移动设备中身份信息采集部件作为外设的装置与方法 |
CN111783594A (zh) * | 2020-06-23 | 2020-10-16 | 杭州海康威视数字技术股份有限公司 | 一种报警方法、装置及电子设备 |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1291763A (zh) * | 2000-05-31 | 2001-04-18 | 成都银晨网讯科技有限公司 | 面像识别ic卡个人身份识别认证系统 |
WO2004003832A2 (en) * | 2002-06-27 | 2004-01-08 | Koninklijke Philips Electronics N.V. | Face recognition using evolutionary algorithms |
CN1794266A (zh) * | 2005-12-31 | 2006-06-28 | 清华大学 | 生物特征融合的身份识别和认证方法 |
JP2006221378A (ja) * | 2005-02-09 | 2006-08-24 | Smart Wireless Kk | 顔認証を利用した警報システムおよび方法 |
WO2006086843A1 (en) * | 2005-02-16 | 2006-08-24 | Orica Explosives Technology Pty Ltd | Blasting methods and apparatus with reduced risk of inadvertent or illicit use |
CN1833397A (zh) * | 2003-06-17 | 2006-09-13 | 联合安全应用Id有限公司 | 用于监视和记录与人有关的活动和数据的电子安全系统 |
CN101102494A (zh) * | 2007-07-23 | 2008-01-09 | 武汉大学 | 一种视频转码中宏块级码率控制方法 |
CN101226653A (zh) * | 2007-01-18 | 2008-07-23 | 中国科学院自动化研究所 | 基于身份证和生物特征识别技术的快速登机系统及方法 |
WO2008088203A1 (es) * | 2006-11-24 | 2008-07-24 | Global Sight, S.A. De C.V. | Sistemas de transmisión de datos en forma remota y digital y localización satelital desde terminales móviles o fijas con cámaras de vigilancia urbana para reconocimiento facial, detector de disparos, captura de personal de seguridad publica y personas extraviadas o secuestradas y alarma ciudadana, vehículos robados, aplicac |
-
2008
- 2008-09-16 WO PCT/CN2008/072367 patent/WO2010031213A1/zh active Application Filing
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1291763A (zh) * | 2000-05-31 | 2001-04-18 | 成都银晨网讯科技有限公司 | 面像识别ic卡个人身份识别认证系统 |
WO2004003832A2 (en) * | 2002-06-27 | 2004-01-08 | Koninklijke Philips Electronics N.V. | Face recognition using evolutionary algorithms |
CN1833397A (zh) * | 2003-06-17 | 2006-09-13 | 联合安全应用Id有限公司 | 用于监视和记录与人有关的活动和数据的电子安全系统 |
JP2006221378A (ja) * | 2005-02-09 | 2006-08-24 | Smart Wireless Kk | 顔認証を利用した警報システムおよび方法 |
WO2006086843A1 (en) * | 2005-02-16 | 2006-08-24 | Orica Explosives Technology Pty Ltd | Blasting methods and apparatus with reduced risk of inadvertent or illicit use |
CN1794266A (zh) * | 2005-12-31 | 2006-06-28 | 清华大学 | 生物特征融合的身份识别和认证方法 |
WO2008088203A1 (es) * | 2006-11-24 | 2008-07-24 | Global Sight, S.A. De C.V. | Sistemas de transmisión de datos en forma remota y digital y localización satelital desde terminales móviles o fijas con cámaras de vigilancia urbana para reconocimiento facial, detector de disparos, captura de personal de seguridad publica y personas extraviadas o secuestradas y alarma ciudadana, vehículos robados, aplicac |
CN101226653A (zh) * | 2007-01-18 | 2008-07-23 | 中国科学院自动化研究所 | 基于身份证和生物特征识别技术的快速登机系统及方法 |
CN101102494A (zh) * | 2007-07-23 | 2008-01-09 | 武汉大学 | 一种视频转码中宏块级码率控制方法 |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106294246A (zh) * | 2016-09-30 | 2017-01-04 | 广州像素数据技术股份有限公司 | 一种移动设备中身份信息采集部件作为外设的装置与方法 |
CN106294246B (zh) * | 2016-09-30 | 2023-09-01 | 广州像素数据技术股份有限公司 | 一种身份信息采集部件作为外设的移动设备与方法 |
CN111783594A (zh) * | 2020-06-23 | 2020-10-16 | 杭州海康威视数字技术股份有限公司 | 一种报警方法、装置及电子设备 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104573679B (zh) | 监控场景下基于深度学习的人脸识别系统 | |
CN110569720A (zh) | 一种基于音视频处理系统的音视频智能识别处理方法 | |
CN106156688A (zh) | 一种动态人脸识别方法及系统 | |
WO2014139416A1 (zh) | 一种突发异常事件智能识别报警装置及报警系统 | |
CN209543514U (zh) | 基于人脸识别的监控与报警系统 | |
CN111126219A (zh) | 一种基于人工智能的变电站人员的身份识别系统和方法 | |
CN112530116A (zh) | 一种基于人脸识别的校园越界自动预警方法 | |
CN103544757A (zh) | 一种门禁控制方法 | |
CN109118617A (zh) | 一种应用于变电站的门禁系统及其识别方法 | |
CN108960107A (zh) | 一种小微型面包车超员识别方法及装置 | |
CN112149638A (zh) | 基于多模态生物特征的人员身份识别系统构建及使用方法 | |
CN106447861A (zh) | 一种高效智能门禁管理系统 | |
CN106652291A (zh) | 一种基于Kinect的室内简易监控报警系统及方法 | |
CN112149576A (zh) | 一种基于图像分析的电梯安全实时监测管理系统 | |
CN107516076A (zh) | 人像识别方法及装置 | |
CN108564704A (zh) | 一种智慧社区单元门 | |
CN111667599A (zh) | 一种人脸识别打卡系统及方法 | |
CN111724526A (zh) | 高铁远程管控作业门 | |
CN208284002U (zh) | 一种智慧社区单元门 | |
CN111385440A (zh) | 具有人脸记录与查询功能的监控摄像机 | |
WO2010031213A1 (zh) | 一种智能身份识别方法及其实现该方法的识别终端 | |
CN214151811U (zh) | 一种多角度人脸识别通行系统 | |
CN113362545A (zh) | 一种声音及视频相结合的居民小区监控系统 | |
CN116416665A (zh) | 基于安防系统的人脸识别方法、装置及存储介质 | |
CN116229624A (zh) | 一种人脸识别智慧门禁疫情防控管理方法及系统 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 08876954 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: 1205A "NOTING OF LOSS OF RIGHTS" DATED 22.09.11 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 08876954 Country of ref document: EP Kind code of ref document: A1 |