WO2020147346A1 - 图像识别方法、系统及装置 - Google Patents

图像识别方法、系统及装置 Download PDF

Info

Publication number
WO2020147346A1
WO2020147346A1 PCT/CN2019/110673 CN2019110673W WO2020147346A1 WO 2020147346 A1 WO2020147346 A1 WO 2020147346A1 CN 2019110673 W CN2019110673 W CN 2019110673W WO 2020147346 A1 WO2020147346 A1 WO 2020147346A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
fused
target object
photographing
similarity
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/110673
Other languages
English (en)
French (fr)
Inventor
李怀奇
冯丽贞
朱凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Gree Electric Appliances Inc of Zhuhai
Original Assignee
Gree Electric Appliances Inc of Zhuhai
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Gree Electric Appliances Inc of Zhuhai filed Critical Gree Electric Appliances Inc of Zhuhai
Publication of WO2020147346A1 publication Critical patent/WO2020147346A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/66—Substation equipment, e.g. for use by subscribers with means for preventing unauthorised or fraudulent calling
    • H04M1/667—Preventing unauthorised calls from a telephone set
    • H04M1/67—Preventing unauthorised calls from a telephone set by electronic means

Definitions

  • This application relates to the field of smart home appliances, and specifically to an image recognition method, system and device.
  • the method of unlocking mobile phones based on face recognition is mostly implemented by using 3D face recognition technology based on the comparison of facial feature points or feature line lengths.
  • This method requires complex face recognition algorithms to achieve face recognition and unlocking mobile phones, and there are calculations.
  • Large quantities, slow unlocking speed, and unrecognizable faces are not negligible shortcomings, while 2D face recognition technology has the advantages of small calculations and fast recognition, but because of the characteristics of too simple judgment, when others use photos, etc.
  • the plane image is easy to be misjudged, so it is easy to reduce the accuracy and safety of recognition.
  • Microwave imaging technology can avoid the problem of misjudgment of images, but the imaging resolution is low.
  • the embodiments of the present application provide an image recognition method, system, and device to at least solve the technical problem of low recognition accuracy when the human face is partially blocked by a 3D recognition technology when the human face is recognized based on the 3D recognition technology.
  • an image recognition method including: acquiring characteristic information of a target object; determining whether the target object is a living body according to the characteristic information; if so, based on at least two camera pairs The target object is photographed to obtain images to be fused corresponding to the at least two photographing devices, wherein the at least two photographing devices include at least a microwave photographing device; the images to be fused are fused to obtain a fusion Image; find whether there is a standard image in the database whose similarity to the fused image exceeds a preset threshold, and if so, the recognition is successful.
  • the characteristic information includes: water content information; acquiring characteristic information of the target object includes: photographing the target object using a microwave photographing device, and determining the scattering field information of the target object indicated by the photographing result The water content information of the target object.
  • determining whether the target object is a living body according to the characteristic information includes: determining whether the water content corresponding to the water content information belongs to a preset threshold interval, and if so, determining that the target object is a living body.
  • the at least two photographing devices further include: a visible light photographing device; photographing the target object based on the at least two photographing devices, and obtaining images to be fused corresponding to the at least two photographing devices respectively includes: The microwave photographing device photographs the target object to obtain a first image to be fused; and based on the visible light photographing device to photograph the target object to obtain a second image to be fused.
  • the method further includes: preprocessing the first image to be fused and/or the second image to be fused, and the preprocessing includes: Binarization; matching the pixel coordinates of the first image to be fused with the second image to be fused.
  • the method before fusing the to-be-fused image to obtain the fused image, the method further includes: comparing the first to-be-fused image with a first standard image, and obtaining the first to-be-fused image and the The first similarity of the first standard image; compare the second to-be-fused image with a second standard image, and obtain the second similarity of the second to-be-fused image and the second standard image; The first similarity and the second similarity determine the weight of the first image to be fused and the second image to be fused when the image to be fused is fused.
  • the weight of the first image to be fused and the second image to be fused includes: If the first degree of similarity is greater than the second degree of similarity, it is determined that when fusing the image to be fused, the weight of the first image to be fused is greater than the weight of the second image to be fused; if the first similarity is If it is less than the second similarity, it is determined that when the image to be fused is fused, the weight of the first image to be fused is smaller than the weight of the second image to be fused.
  • an image recognition system including: a microwave imaging device to obtain characteristic information of a target object; a recognition device to determine whether the target object is a living body according to the characteristic information; if so, The target object is photographed based on at least two kinds of photographing devices to obtain images to be fused corresponding to the at least two kinds of photographing devices, wherein the at least two photographing devices include at least the microwave photographing device; The image to be fused is fused to obtain a fused image; whether there is a standard image in the database whose similarity to the fused image exceeds a preset threshold, and if there is, the recognition is successful.
  • a household appliance including: a microwave imaging device to obtain characteristic information of a target object; a processor to determine whether the target object is a living body according to the characteristic information; , The target object is photographed based on at least two photographing devices to obtain images to be fused corresponding to the at least two photographing devices, wherein the at least two photographing devices include at least the microwave photographing device; The image to be fused is fused to obtain a fused image; whether there is a standard image in the database whose similarity to the fused image exceeds a preset threshold, and if there is, the recognition is successful.
  • an image recognition device including: an acquisition module configured to acquire characteristic information of a target object; a determining module configured to determine whether the target object is a living body according to the characteristic information; If so, the target object is photographed based on at least two photographing devices to obtain images to be fused corresponding to the at least two photographing devices, wherein the at least two photographing devices include at least a microwave photographing device; processing The module is configured to fuse the image to be fused to obtain a fused image; the recognition module is configured to find whether there is a standard image in the database whose similarity to the fused image exceeds a preset threshold, and if there is, the recognition is successful .
  • a storage medium includes a stored program, and the device where the storage medium is located is controlled to execute the above-mentioned image recognition method when the program runs.
  • a processor configured to run a program, wherein the image recognition method described above is executed when the program is running.
  • an image recognition method which includes: photographing a target object based on a microwave photographing device to obtain a first to-be-fused image; photographing the target object based on a visible light photographing device to obtain a first Two images to be fused; the first image to be fused and the second image to be fused to obtain a fused image; to find whether there is a standard image in the database whose similarity to the fused image exceeds a preset threshold, if any , The recognition is successful.
  • the characteristic information of the target object is acquired; whether the target object is a living body is determined according to the characteristic information; if so, the target object is photographed based on at least two photographing devices, and the difference between the target object and the The images to be fused corresponding to the at least two photographing devices, wherein the at least two photographing devices include at least a microwave photographing device; the images to be fused are merged to obtain the fused image; The standard image whose similarity of the fused image exceeds the preset threshold. If there is, the recognition is successful.
  • the microwave imaging technology with the visible light imaging technology
  • Fig. 1 is a schematic flowchart of an optional image recognition method according to an embodiment of the present application
  • Fig. 2 is a schematic flowchart of an optional image recognition method according to an embodiment of the present application
  • Fig. 3 is a schematic structural diagram of an optional image recognition system according to an embodiment of the present application.
  • Figure 4 is a schematic structural diagram of an optional household appliance according to an embodiment of the present application.
  • Fig. 5 is a schematic structural diagram of an optional image recognition device according to an embodiment of the present application.
  • Fig. 6 is a flowchart of another image recognition method according to an embodiment of the present application.
  • an embodiment of an image recognition method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although The logical sequence is shown in the flowchart, but in some cases, the steps shown or described may be performed in a different order than here.
  • Fig. 1 is a schematic flowchart of an image recognition method according to an embodiment of the present application. As shown in Fig. 1, the method at least includes the following steps:
  • Step S102 acquiring characteristic information of the target object
  • the target object is a target user whose identity is to be identified
  • the characteristic information may be: water content information
  • acquiring characteristic information of the target object may be achieved in the following manner: using a microwave imaging device to photograph the target object, and based on the scattered field information of the target object indicated by the shooting result Determine the water content information of the target object.
  • the foregoing scattered field information may be energy information fed back to the microwave radar after the target object is photographed by the microwave imaging device.
  • the water content of the target object is different, the energy information feedback is different.
  • Step S104 Determine whether the target object is a living body according to the characteristic information
  • determining whether the target object is a living body according to the characteristic information can be achieved in the following manner: determining whether the water content corresponding to the water content information belongs to a preset threshold interval, if yes , It is determined that the target object is a living body.
  • the human body or living body contains water, and the water content is different from ordinary objects.
  • the preset threshold interval may be 70% to 80%
  • the microwave imaging device may be a microwave radar.
  • the microwave imaging device when the microwave imaging device detects an object nearby, it can identify whether the object is alive by detecting water content and the heart rate of the target object.
  • Step S106 If yes, the target object is photographed based on at least two kinds of photographing devices to obtain images to be fused corresponding to the at least two kinds of photographing devices, wherein the at least two kinds of photographing devices include at least microwave photography Device
  • the at least two types of imaging devices further include visible light imaging devices; the target object is captured based on the at least two types of imaging devices, and the results correspond to the at least two types of imaging devices respectively.
  • the image to be fused includes: photographing the target object based on the microwave imaging device to obtain a first image to be fused; and photographing the target object based on the visible light imaging device to obtain a second image to be fused.
  • the visible light photographing device and the microwave photographing device may be a front camera device of the terminal; or a rear camera device of the terminal.
  • the above-mentioned visible light photographing device is based on 3D photographing technology when photographing the target object.
  • Step S108 fusing the image to be fused to obtain a fused image
  • the method before fusing the image to be fused to obtain the fused image, the method further needs to perform the following steps: performing the following steps on the first image to be fused and/or the second image to be fused Image preprocessing is performed, and the preprocessing includes: binarization; performing pixel coordinate matching on the first image to be fused and the second image to be fused.
  • the above-mentioned first image to be fused is a 3D image
  • the binarization is used in the 3D image taken from the visible light camera and the second to be fused image taken from the microwave camera.
  • the feature points and feature lines of the face are obtained.
  • the pixel coordinate matching of the first to-be-fused image and the second to-be-fused image can be achieved in the following manner: based on the feature points and feature lines of the target object being Based on the principle that the coordinates of the pixels in the first to-be-fused image correspond to those in the second to-be-fused image, the first to-be-fused image and the second to-be-fused image are cut or converted into images with the same number of pixels.
  • step S110 it is searched whether there is a standard image whose similarity with the fused image exceeds a preset threshold in the database, and if there is, the recognition is successful.
  • a standard image of the target object is pre-stored in the database, and the standard image is a standard image obtained by fusing a 3D image taken by a visible light camera and an image taken by a microwave camera. After fusing the first image to be fused with the second image to be fused, compare with each image in the database, or compare a group of images, when it is found that there is a standard that the similarity with the fused image exceeds the preset threshold in the database When displaying images, it means that the current target object has been granted the qualification to unlock the terminal, or the qualification to obtain the information in the terminal, etc., then it is determined that the image information of the current target object is stored in the database, that is, the recognition is successful.
  • the database is a cloud database or the internal memory of the terminal. .
  • the following steps need to be performed: comparing the first image to be fused with the first standard image, and obtaining all the images.
  • the second degree of similarity; the weight of the first image to be fused and the second image to be fused when the image to be fused is determined based on the first similarity and the second similarity.
  • the database also stores a first standard image of the target object captured only with a visible light camera, and a second standard image captured only with a microwave camera.
  • the first image to be fused and the second The weight of the fused image can be realized in the following manner: if the first similarity is greater than the second similarity, it is determined that when the image to be fused is fused, the weight of the first image to be fused is greater than the second similarity. The weight of the image to be fused; if the first similarity is less than the second similarity, it is determined that when the image to be fused is fused, the weight of the first image to be fused is smaller than the weight of the second image to be fused .
  • the weight of the first image to be fused is set to be greater than the weight of the second image to be fused.
  • the first similarity can be determined based on the ratio of the first similarity to the second similarity.
  • the weight of the image to be fused and the weight of the second image to be fused for example, the ratio of the weight of the first image to be fused to the weight of the second image to be fused can be set to be the same as the ratio of the first similarity to the second similarity. For example: if the first degree of similarity is 90% and the second degree of similarity is 30%, when fusing the images to be fused, set the weight of the first image to be fused to 0.75 and the weight of the second image to be fused to 0.25.
  • microwave imaging and visible light imaging technology by adopting microwave imaging and visible light imaging technology, it combines the advantages that microwave can pass through obstacles such as masks and sunglasses to capture facial features, and visible light can capture facial details with higher resolution. Improve the accuracy of face recognition.
  • the characteristic information of the target object is acquired; whether the target object is a living body is determined according to the characteristic information; if so, the target object is photographed based on at least two photographing devices, and the difference between the target object and the The images to be fused corresponding to the at least two photographing devices, wherein the at least two photographing devices include at least a microwave photographing device; the images to be fused are merged to obtain the fused image; The standard image whose similarity of the fused image exceeds the preset threshold. If there is, the recognition is successful.
  • the microwave imaging technology with the visible light imaging technology
  • Fig. 2 is a schematic flowchart of an image recognition method according to an embodiment of the present application. As shown in Fig. 2, the method at least includes the following steps:
  • Step S202 the microwave radar imaging module obtains the water content of the measured object
  • a microwave radar imaging module is used to photograph the target object, and the water content information of the target object is determined based on the scattered field information of the target object indicated by the photographing result.
  • the foregoing scattered field information may be energy information fed back to the microwave radar after the target object is photographed by the microwave radar imaging module.
  • the water content of the target object is different, the energy information feedback is different.
  • the target object may be the user to be identified.
  • Step S204 judging whether the measured object is a living body
  • determining whether the measured object is a living body can be achieved in the following manner: determining whether the water content corresponding to the water content information belongs to a preset threshold interval, and if so, determining the target The object is a living body.
  • the human body or living body contains water, and the water content is different from ordinary objects.
  • the preset threshold interval may be 70% to 80%.
  • Step S206 if yes, the visible light camera module captures the face image, and the microwave radar imaging module obtains the face image.
  • step S208 the images are merged.
  • the visible light camera module captures a face image
  • the microwave radar imaging module obtains the face image
  • the face image captured by the visible light camera module is combined with the face image obtained by the microwave radar imaging module Perform fusion to obtain the fused image.
  • the following steps need to be performed: preprocessing the face image taken by the visible light camera module and/or the face image obtained by the microwave radar imaging module, so The preprocessing includes: binarization; performing pixel coordinate matching between the face image taken by the visible light camera module and the face image obtained by the microwave radar imaging module.
  • the face image taken by the visible light camera module is a 3D image
  • the binarization is used in the 3D image taken from the visible light camera module and the microwave taken from the microwave imaging device. From the face image acquired by the radar imaging module, the feature points and feature lines of the face are acquired.
  • the pixel coordinate matching of the face image captured by the visible light camera module with the face image acquired by the microwave radar imaging module can be achieved in the following manner: The principle that the feature points and feature lines in the face image captured by the visible light camera module correspond to the pixel coordinates in the face image obtained by the microwave radar imaging module, the face image captured by the visible light camera module and the microwave radar imaging module are obtained Cut or convert the face image into an image with the same number of pixels.
  • Step S210 compare the similarity with existing face data in the database
  • the fused image is combined with the existing face data in the database. Similarity comparison.
  • step S212 it is determined whether the similarity is higher than a preset threshold, if so, step S214 is executed, if not, step S212 is executed, or the recognition failure is prompted.
  • a standard image of the target object is pre-stored in the database, and the standard image is a standard image obtained by fusing the 3D image taken by the visible light camera module and the image taken by the microwave radar imaging module. After fusing the face image taken by the visible light camera module with the face image obtained by the microwave radar imaging module, compare it with each image in the database, or compare a group of images, when the database is found to be similar to the fused image.
  • the standard image whose degree exceeds the preset threshold, it means that the current target object has been granted the qualification to unlock the terminal, or the qualification to obtain the information in the terminal, etc., then it is determined that the image information of the current target object is stored in the database, that is, the recognition is successful. It is a cloud database or terminal internal storage.
  • Step S214 it is determined that the recognition is successful.
  • Fig. 3 is a schematic structural diagram of an image recognition system according to an embodiment of the present application. As shown in Fig. 3, the system at least includes: a microwave imaging device 32 and a recognition device 34; among them:
  • the microwave imaging device 32 obtains characteristic information of the target object
  • the recognition device 34 determines whether the target object is a living body according to the characteristic information; if so, the target object is photographed based on at least two photographing devices to obtain the to-be-fused images corresponding to the at least two photographing devices respectively , wherein the at least two photographing devices include at least the microwave photographing device; fusing the image to be fused to obtain a fused image; searching for whether there is a similarity with the fused image in the database that exceeds a preset threshold If there is a standard image, the recognition is successful.
  • Fig. 4 is a schematic structural diagram of a household appliance according to an embodiment of the present application. As shown in Fig. 4, the household appliance at least includes: a microwave imaging device 42 and a processor 44; wherein:
  • the microwave imaging device 42 obtains characteristic information of the target object
  • the processor 44 determines whether the target object is a living body according to the characteristic information; if so, the target object is photographed based on at least two kinds of photographing devices to obtain images to be fused corresponding to the at least two kinds of photographing devices respectively , wherein the at least two photographing devices include at least the microwave photographing device; fusing the image to be fused to obtain a fused image; searching for whether there is a similarity with the fused image in the database that exceeds a preset threshold If there is a standard image, the recognition is successful.
  • Fig. 5 is a schematic structural diagram of an image recognition device according to an embodiment of the present application. As shown in Fig. 5, the device at least includes: an acquisition module 52, a determination module 54, a processing module 56, and an identification module 58; among them:
  • the obtaining module 52 is configured to obtain characteristic information of the target object
  • the determining module 54 is configured to determine whether the target object is a living body according to the characteristic information; if so, the target object is photographed based on at least two types of photographing devices, and the standby corresponding to the at least two types of photographing devices are obtained. Fused images, wherein the at least two types of imaging devices include at least a microwave imaging device;
  • the processing module 56 is configured to fuse the image to be fused to obtain a fused image
  • the recognition module 58 is configured to search the database for whether there is a standard image whose similarity to the fused image exceeds a preset threshold. If there is, the recognition is successful.
  • Fig. 6 is a flowchart of another image recognition method according to an embodiment of the present application. As shown in Figure 6, the method includes:
  • Step S602 photographing the target object based on the microwave photographing device to obtain the first image to be fused
  • Step S604 shooting the target object based on the visible light shooting device to obtain a second to-be-fused image
  • Step S606 fusing the first image to be fused and the second image to be fused to obtain a fused image
  • step S608 it is searched whether there is a standard image whose similarity to the fused image exceeds a preset threshold in the database, and if there is, the recognition is successful.
  • step S604 may be executed first, and then step S602 may be executed.
  • a storage medium characterized in that the storage medium includes a stored program, wherein the device where the storage medium is located is controlled to execute the aforementioned image recognition method when the program runs.
  • a processor characterized in that the processor is used to run a program, wherein the above-mentioned image recognition method is executed when the program is running.
  • the disclosed technical content may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of units may be a logical function division.
  • multiple units or components may be combined or integrated into Another system, or some features can be ignored, or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, units or modules, and may be in electrical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or software function unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.
  • the technical solution of the present application essentially or part of the contribution to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to enable a computer device (which may be a personal computer, server, network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application.
  • the aforementioned storage media include: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code .
  • the solution provided by the embodiment of this application is applied to the image recognition process.
  • the solution provided by the embodiment of this application can be used to obtain a fusion image based on two shooting devices when the target object is blocked.
  • the technology recognizes the human face, if the human face is partially obscured by an obstruction, the recognition accuracy is low.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Image Analysis (AREA)

Abstract

一种图像识别方法、系统及装置。其中，该方法包括：获取目标对象的特征信息(S102)；根据所述特征信息确定所述目标对象是否为活体(S104)；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像(S106)，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；将所述待融合图像进行融合，得到融合图像(S108)；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功(S110)。

Description

图像识别方法、系统及装置
本申请要求于2019年1月15日提交至中国国家知识产权局、申请号为201910036596.3、发明名称为“图像识别方法、系统及装置”的专利申请的优先权。
技术领域
本申请涉及智能家电领域，具体而言，涉及一种图像识别方法、系统及装置。
背景技术
目前，基于人脸识别解锁手机的方式，多采用3D人脸识别技术基于面部特征点或特征线长度对比实现，该方法需要通过复杂的人脸识别算法来实现人脸识别解锁手机，存在着运算量大，解锁速度慢，人脸被遮挡无法识别等不可忽视的缺点，而2D人脸识别技术虽然具有运算量小、识别快的优点，但是因为存在判定过于简单的特点，当他人使用照片等平面图像时容易被误判，所以容易降低识别的精准性和安全性。微波成像技术可以避免误判图像的问题，但是成像的分辨率较低。
针对上述问题，目前尚未提出有效的解决方案。
发明内容
本申请实施例提供了一种图像识别方法、系统及装置，以至少解决基于3D识别技术对人脸进行识别时，若人脸被部分遮挡物遮挡，识别准确率较低的技术问题。
根据本申请实施例的一个方面，提供了一种图像识别方法，包括：获取目标对象的特征信息；根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
可选地，所述特征信息包括：含水量信息；获取目标对象的特征信息包括：使用微波拍摄装置对所述目标对象进行拍摄，基于拍摄结果指示的所述目标对象的散射场信息确定所述目标对象的含水量信息。
可选地，根据所述特征信息确定所述目标对象是否为活体包括：判断所述含水量 信息对应的含水量是否属于预设阈值区间，若是，则确定所述目标对象为活体。
可选地，所述至少两种拍摄装置还包括：可见光拍摄装置；基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像包括：基于所述微波拍摄装置对所述目标对象进行拍摄，得到第一待融合图像；基于所述可见光拍摄装置对所述目标对象进行拍摄，得到第二待融合图像。
可选地，将所述待融合图像进行融合，得到融合图像之前，所述方法还包括：对所述第一待融合图像和/或第二待融合图像进行预处理，所述预处理包括：二值化；将所述第一待融合图像与所述第二待融合图像进行像素坐标匹配。
可选地，将所述待融合图像进行融合，得到融合图像之前，所述方法还包括：将所述第一待融合图像与第一标准图像进行对比，获取所述第一待融合图像与所述第一标准图像的第一相似度；将所述第二待融合图像与第二标准图像进行对比，获取所述第二待融合图像与所述第二标准图像的第二相似度；基于所述第一相似度与所述第二相似度确定对所述待融合图像进行融合时，所述第一待融合图像与所述第二待融合图像的权重。
可选地，基于所述第一相似度与所述第二相似度确定对所述待融合图像进行融合时，所述第一待融合图像与所述第二待融合图像的权重包括：若所述第一相似度大于所述第二相似度，则确定对所述待融合图像进行融合时，第一待融合图像的权重大于所述第二待融合图像的权重；若所述第一相似度小于所述第二相似度，则确定对所述待融合图像进行融合时，第一待融合图像的权重小于所述第二待融合图像的权重。
根据本申请实施例的一个方面，提供了一种图像识别系统，包括：微波拍摄装置，获取目标对象的特征信息；识别设备，根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括所述微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
根据本申请实施例的一个方面，提供了一种家用电器，该家用电器包括：微波拍摄装置，获取目标对象的特征信息；处理器，根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括所述微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
根据本申请实施例的一个方面，提供了一种图像识别装置，包括：获取模块，设置为获取目标对象的特征信息；确定模块，设置为根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；处理模块，设置为将所述待融合图像进行融合，得到融合图像；识别模块，设置为查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
根据本申请实施例的一个方面，提供了一种存储介质，所述存储介质包括存储的程序，其中，在所述程序运行时控制所述存储介质所在设备执行上述的图像识别方法。
根据本申请实施例的一个方面，提供了一种处理器，所述处理器设置为运行程序，其中，所述程序运行时执行上述的图像识别方法。
根据本申请实施例的一个方面，提供了一种图像识别方法，包括：基于微波拍摄装置对目标对象进行拍摄，得到第一待融合图像；基于可见光拍摄装置对所述目标对象进行拍摄，得到第二待融合图像；将所述第一待融合图像和第二待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
在本申请实施例中，通过获取目标对象的特征信息；根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功的方式，达到了通过将微波成像技术与可见光成像技术结合，将两者的优势互补，提高在人脸识别过程中，人脸被部分遮挡的情况下，对人脸识别的精度的技术效果，从而解决了基于3D识别技术对人脸进行识别时，若人脸被部分遮挡物遮挡，识别准确率较低的技术问题。
附图说明
此处所说明的附图用来提供对本申请的进一步理解，构成本申请的一部分，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。在附图中：
图1是根据本申请实施例的一种可选的图像识别的方法的流程示意图；
图2是根据本申请实施例的一种可选的图像识别的方法的流程示意图；
图3是根据本申请实施例的一种可选的图像识别的系统的结构示意图；
图4是根据本申请实施例的一种可选的家用电器的结构示意图；
图5是根据本申请实施例的一种可选的图像识别的装置的结构示意图；
图6是根据本申请实施例的另一种图像识别的方法的流程图。
具体实施方式
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分的实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本申请保护的范围。
需要说明的是，本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
根据本申请实施例，提供了一种图像识别的方法实施例，需要说明的是，在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行，并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
图1是根据本申请实施例的图像识别的方法的流程示意图，如图1所示，该方法至少包括如下步骤：
步骤S102，获取目标对象的特征信息；
在本申请的一些可选的实施例中，目标对象为待识别身份的目标用户，所述特征信息可以为：含水量信息；
在本申请的一些可选的实施例中，获取目标对象的特征信息可以通过以下方式进行实现：使用微波拍摄装置对所述目标对象进行拍摄，基于拍摄结果指示的所述目标 对象的散射场信息确定所述目标对象的含水量信息。
在本申请的一些可选的实施例中，上述散射场信息可以为目标对象在所述微波拍摄装置拍摄后，反馈到微波雷达的能量信息。目标对象的含水量不同，反馈的能量信息不同。
步骤S104，根据所述特征信息确定所述目标对象是否为活体；
在本申请的一些可选的实施例中，根据所述特征信息确定所述目标对象是否为活体可以通过以下方式进行实现：判断所述含水量信息对应的含水量是否属于预设阈值区间，若是，则确定所述目标对象为活体。
人体或者活体含有水分，含水量与普通物体不同，在本申请的一些可选的实施例中，上述预设阈值区间可以为70％至80％，其中，微波拍摄装置可以为微波雷达。
在本申请的其他一些可选的实施例中，当微波拍摄装置，探测到有物体在附近时，通过检测含水量和与目标对象的心率等方式识别该物体是否为活体。
步骤S106，若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；
在本申请的一些可选的实施例中，所述至少两种拍摄装置还包括可见光拍摄装置；基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像包括：基于所述微波拍摄装置对所述目标对象进行拍摄，得到第一待融合图像；基于所述可见光拍摄装置对所述目标对象进行拍摄，得到第二待融合图像。
在本申请的一些可选的实施例中，可见光拍摄装置与微波拍摄装置可以为终端前置摄像装置；也可以为终端的后置摄像装置。
在本申请的一些可选的实施例中，上述可见光拍摄装置对目标对象进行拍摄时，基于3D拍摄技术。
步骤S108，将所述待融合图像进行融合，得到融合图像；
在本申请的一些可选的实施例中，将所述待融合图像进行融合，得到融合图像之前，所述方法还需执行以下步骤：对所述第一待融合图像和/或第二待融合图像进行预处理，所述预处理包括：二值化；将所述第一待融合图像与所述第二待融合图像进行像素坐标匹配。
在本申请的一些可选的实施例中，上述第一待融合图像为3D图像，所述二值化用于从可见光拍摄装置拍摄的3D图像中，以及从微波拍摄装置拍摄的第二待融合图像中，获取人脸的特征点与特征线。
在本申请的一些可选的实施例中，将所述第一待融合图像与所述第二待融合图像进行像素坐标匹配可通过以下方式进行实现：基于使目标对象的特征点与特征线在第一待融合图像中与第二待融合图像中的像素坐标相对应的原则，将第一待融合图像与第二待融合图像剪切或转换为像素数目相同的图像。
步骤S110，查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
在本申请的一些可选的实施例中，数据库中预先存储有目标对象的标准图像，该标准图像为可见光拍摄装置拍摄的3D图像与微波拍摄装置拍摄的图像融合后的标准图像。在将第一待融合图像与第二待融合图像进行融合后，与数据库中的各个图像对比，或某组图像对比，当查找到数据库中存在与融合图像的相似度超过预设阈值的标准图像时，表示当前目标对象已被授予终端开锁资格，或获取终端中信息的资格等，则确定数据库中存储有当前目标对象的图像信息，即识别成功，其中，数据库为云端数据库或者终端内部存储器。
在本申请的一些可选的实施例中，将所述待融合图像进行融合，得到融合图像之前，还需执行以下步骤：将所述第一待融合图像与第一标准图像进行对比，获取所述第一待融合图像与所述第一标准图像的第一相似度；将所述第二待融合图像与第二标准图像进行对比，获取所述第二待融合图像与所述第二标准图像的第二相似度；基于所述第一相似度与所述第二相似度确定对所述待融合图像进行融合时，所述第一待融合图像与所述第二待融合图像的权重。
在本申请的一些可选的实施例中，数据库中还存储有仅用可见光拍摄装置拍摄的目标对象的第一标准图像，以及仅用微波拍摄装置拍摄的第二标准图像。
在本申请的一些可选的实施例中，基于所述第一相似度与所述第二相似度确定对所述待融合图像进行融合时，所述第一待融合图像与所述第二待融合图像的权重可以通过以下方式进行实现：若所述第一相似度大于所述第二相似度，则确定对所述待融合图像进行融合时，第一待融合图像的权重大于所述第二待融合图像的权重；若所述第一相似度小于所述第二相似度，则确定对所述待融合图像进行融合时，第一待融合图像的权重小于所述第二待融合图像的权重。
在本申请的一些可选的实施例中，设定第一待融合图像的权重大于第二待融合图 像的权重，具体的，可以基于第一相似度与第二相似度的比值，确定第一待融合图像的权重与第二待融合图像的权重，例如：可设定第一待融合图像的权重与第二待融合图像的权重的比值和第一相似度与第二相似度的比值相同。例如：若第一相似度为90％，第二相似度为30％，则在对待融合图像进行融合时，设定第一待融合图像的权重为0.75，第二待融合图像的权重为0.25。
本申请的方案中，通过采用微波成像以及可见光成像技术，结合了微波可穿过口罩、墨镜等障碍物拍摄到人脸特征，以及可见光可拍摄到人脸细节，分辨率较高的优点，提高了人脸识别的准确度。
在本申请实施例中，通过获取目标对象的特征信息；根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功的方式，达到了通过将微波成像技术与可见光成像技术结合，将两者的优势互补，提高在人脸识别过程中，人脸被部分遮挡的情况下，对人脸识别的精度的技术效果，从而解决了基于3D识别技术对人脸进行识别时，若人脸被部分遮挡物遮挡，识别准确率较低的技术问题。
图2是根据本申请实施例的图像识别的方法的流程示意图，如图2所示，该方法至少包括如下步骤：
步骤S202，微波雷达成像模块获取被测对象含水量；
在本申请的一些可选的实施例中，使用微波雷达成像模块对所述目标对象进行拍摄，基于拍摄结果指示的所述目标对象的散射场信息确定所述目标对象的含水量信息。上述散射场信息可以为目标对象在所述微波雷达成像模块拍摄后，反馈到微波雷达的能量信息。目标对象的含水量不同，反馈的能量信息不同。其中，目标对象可以为待识别用户。
步骤S204，判断被测对象是否为活体；
在本申请的一些可选的实施例中，判断被测对象是否为活体可通过以下方式进行实现：判断所述含水量信息对应的含水量是否属于预设阈值区间，若是，则确定所述目标对象为活体。
人体或者活体含有水分，含水量与普通物体不同，在本申请的一些可选的实施例中，上述预设阈值区间可以为70％至80％。
步骤S206，若是，则可见光摄像头模块拍摄人脸图像，微波雷达成像模块获取人脸图像。
步骤S208，融合图像。
在本申请的一些可选的实施例中，可见光摄像头模块拍摄人脸图像，微波雷达成像模块获取人脸图像之后，将可见光摄像头模块拍摄的人脸图像，与微波雷达成像模块获取的人脸图像进行融合，获取到融合后的图像。
在本申请的一些可选的实施例中，融合图像之前，还需执行以下步骤：对所述可见光摄像头模块拍摄的人脸图像和/或微波雷达成像模块获取的人脸图像进行预处理，所述预处理包括：二值化；将所述可见光摄像头模块拍摄的人脸图像与所述微波雷达成像模块获取的人脸图像进行像素坐标匹配。
在本申请的一些可选的实施例中，上述可见光摄像头模块拍摄的人脸图像为3D图像，所述二值化用于从可见光摄像头模块拍摄的3D图像中，以及从微波拍摄装置拍摄的微波雷达成像模块获取的人脸图像中，获取人脸的特征点与特征线。
在本申请的一些可选的实施例中，将所述可见光摄像头模块拍摄的人脸图像与所述微波雷达成像模块获取的人脸图像进行像素坐标匹配可通过以下方式进行实现：基于使目标对象的特征点与特征线在可见光摄像头模块拍摄的人脸图像中与微波雷达成像模块获取的人脸图像中的像素坐标相对应的原则，将可见光摄像头模块拍摄的人脸图像与微波雷达成像模块获取的人脸图像剪切或转换为像素数目相同的图像。
步骤S210，与数据库已有的人脸数据进行相似度比较
在本申请的一些可选的实施例中，将可见光摄像头模块拍摄的人脸图像与微波雷达成像模块获取的人脸图像进行融合之后，将融合后的图像与数据库中已有的人脸数据进行相似度比较。
步骤S212，判断相似度是否高于预设阈值，若是，则执行步骤S214，若否，则执行步骤S212，或提示识别失败。
查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
在本申请的一些可选的实施例中，数据库中预先存储有目标对象的标准图像，该标准图像为可见光摄像头模块拍摄的3D图像与微波雷达成像模块拍摄的图像融合后的标准图像。在将可见光摄像头模块拍摄的人脸图像与微波雷达成像模块获取的人脸图像进行融合后，与数据库中的各个图像对比，或某组图像对比，当查找到数据库中 存在与融合图像的相似度超过预设阈值的标准图像时，表示当前目标对象已被授予终端开锁资格，或获取终端中信息的资格等，则确定数据库中存储有当前目标对象的图像信息，即识别成功，其中，数据库为云端数据库或者终端内部存储器。
步骤S214，确定识别成功。
图3是根据本申请实施例的图像识别系统的结构示意图，如图3所示，该系统至少包括：微波拍摄装置32、识别设备34；其中：
微波拍摄装置32，获取目标对象的特征信息；
识别设备34，根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括所述微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
需要说明的是，图3所示实施例的优选实施方式可以参见图1所示实施例的相关描述，此处不再赘述。
图4是根据本申请实施例的家用电器的结构示意图，如图4所示，该家用电器至少包括：微波拍摄装置42、处理器44；其中：
微波拍摄装置42，获取目标对象的特征信息；
处理器44，根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括所述微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
需要说明的是，图4所示实施例的优选实施方式可以参见图1所示实施例的相关描述，此处不再赘述。
图5是根据本申请实施例的图像识别装置的结构示意图，如图5所示，该装置至少包括：获取模块52、确定模块54、处理模块56、识别模块58；其中：
获取模块52，设置为获取目标对象的特征信息；
确定模块54，设置为根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置 对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；
处理模块56，设置为将所述待融合图像进行融合，得到融合图像；
识别模块58，设置为查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
需要说明的是，图5所示实施例的优选实施方式可以参见图1所示实施例的相关描述，此处不再赘述。
图6是根据本申请实施例的另一种图像识别的方法的流程图。如图6所示，该方法包括：
步骤S602，基于微波拍摄装置对目标对象进行拍摄，得到第一待融合图像；
步骤S604，基于可见光拍摄装置对所述目标对象进行拍摄，得到第二待融合图像；
步骤S606，将所述第一待融合图像和第二待融合图像进行融合，得到融合图像；
步骤S608，查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
其中，步骤S602和步骤S604的执行顺序是可以互换的，即在本申请的一些实施例中，可以先执行步骤S604，然后再执行步骤S602。
需要说明的是，图6所示实施例的优选实施方式可以参见图1所示实施例的相关描述，此处不再赘述。
根据本申请实施例的另一个方面，还提供了一种存储介质，其特征在于，存储介质包括存储的程序，其中，在程序运行时控制存储介质所在设备执行上述的图像识别的方法。
根据本申请实施例的另一个方面，还提供了一种处理器，其特征在于，处理器用于运行程序，其中，程序运行时执行上述的图像识别的方法。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
在本申请的上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的技术内容，可通过其它的方式实现。其中，以上所描述的装置实施例仅仅是示意性的，例如单元的划分，可以为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可 以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，单元或模块的间接耦合或通信连接，可以是电性或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述仅是本申请的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本申请原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本申请的保护范围。
工业实用性
本申请实施例提供的方案应用于图像识别过程中，采用本申请实施例提供的方案，可以使得目标对象被遮挡时，基于两种拍摄装置得到融合图像，从而通过查抄数据库进而解决了基于3D识别技术对人脸进行识别时，若人脸被部分遮挡物遮挡，识别准确率较低的技术问题。

Claims (13)

  1. 一种图像识别方法，包括：
    获取目标对象的特征信息；
    根据所述特征信息确定所述目标对象是否为活体；
    若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；
    将所述待融合图像进行融合，得到融合图像；
    查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
  2. 根据权利要求1所述的方法，其中，所述特征信息包括：含水量信息；
    获取目标对象的特征信息包括：使用微波拍摄装置对所述目标对象进行拍摄，基于拍摄结果指示的所述目标对象的散射场信息确定所述目标对象的含水量信息。
  3. 根据权利要求2所述的方法，其中，根据所述特征信息确定所述目标对象是否为活体包括：
    判断所述含水量信息对应的含水量是否属于预设阈值区间，若是，则确定所述目标对象为活体。
  4. 根据权利要求1所述的方法，其中，所述至少两种拍摄装置还包括：可见光拍摄装置；
    基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像包括：
    基于所述微波拍摄装置对所述目标对象进行拍摄，得到第一待融合图像；
    基于所述可见光拍摄装置对所述目标对象进行拍摄，得到第二待融合图像。
  5. 根据权利要求4所述的方法，其中，将所述待融合图像进行融合，得到融合图像之前，所述方法还包括：
    对所述第一待融合图像和第二待融合图像中的至少之一进行预处理，所述预处理包括：二值化；
    将所述第一待融合图像与所述第二待融合图像进行像素坐标匹配。
  6. 根据权利要求4所述的方法，其中，将所述待融合图像进行融合，得到融合图像之前，所述方法还包括：
    将所述第一待融合图像与第一标准图像进行对比，获取所述第一待融合图像与所述第一标准图像的第一相似度；
    将所述第二待融合图像与第二标准图像进行对比，获取所述第二待融合图像与所述第二标准图像的第二相似度；
    基于所述第一相似度与所述第二相似度确定对所述待融合图像进行融合时，所述第一待融合图像与所述第二待融合图像的权重。
  7. 根据权利要求6所述的方法，其中，基于所述第一相似度与所述第二相似度确定对所述待融合图像进行融合时，所述第一待融合图像与所述第二待融合图像的权重包括：
    若所述第一相似度大于所述第二相似度，则确定对所述待融合图像进行融合时，第一待融合图像的权重大于所述第二待融合图像的权重；
    若所述第一相似度小于所述第二相似度，则确定对所述待融合图像进行融合时，第一待融合图像的权重小于所述第二待融合图像的权重。
  8. 一种图像识别系统，包括：
    微波拍摄装置，获取目标对象的特征信息；
    识别设备，根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括所述微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
  9. 一种家用电器，包括：
    微波拍摄装置，获取目标对象的特征信息；
    处理器，根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括所述微波拍摄装置；将所述待融合图像进行融合，得到融合图像；查找数据库中的是否存在与所述融 合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
  10. 一种图像识别装置，包括：
    获取模块，设置为获取目标对象的特征信息；
    确定模块，设置为根据所述特征信息确定所述目标对象是否为活体；若是，则基于至少两种拍摄装置对所述目标对象进行拍摄，得到分别与所述至少两种拍摄装置对应的待融合图像，其中，所述至少两种拍摄装置中至少包括微波拍摄装置；
    处理模块，设置为将所述待融合图像进行融合，得到融合图像；
    识别模块，设置为查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
  11. 一种存储介质，包括存储的程序，其中，在所述程序运行时控制所述存储介质所在设备执行权利要求1至7中任意一项所述的图像识别方法。
  12. 一种处理器，所述处理器用于运行程序，其中，所述程序运行时执行权利要求1至7中或任意一项所述的图像识别方法。
  13. 一种图像识别方法，包括：
    基于微波拍摄装置对目标对象进行拍摄，得到第一待融合图像；基于可见光拍摄装置对所述目标对象进行拍摄，得到第二待融合图像；
    将所述第一待融合图像和第二待融合图像进行融合，得到融合图像；
    查找数据库中的是否存在与所述融合图像的相似度超过预设阈值的标准图像，若有，则识别成功。
PCT/CN2019/110673 2019-01-15 2019-10-11 图像识别方法、系统及装置 Ceased WO2020147346A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910036596.3 2019-01-15
CN201910036596.3A CN109800699A (zh) 2019-01-15 2019-01-15 图像识别方法、系统及装置

Publications (1)

Publication Number Publication Date
WO2020147346A1 true WO2020147346A1 (zh) 2020-07-23

Family

ID=66559155

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/110673 Ceased WO2020147346A1 (zh) 2019-01-15 2019-10-11 图像识别方法、系统及装置

Country Status (2)

Country Link
CN (1) CN109800699A (zh)
WO (1) WO2020147346A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11295538B2 (en) 2020-02-20 2022-04-05 Asustek Computer Inc. Method and equipment for constructing three-dimensional face model
CN114926717A (zh) * 2022-04-18 2022-08-19 杭州海康威视数字技术股份有限公司 对象特征融合方法、装置以及存储介质
CN117876429A (zh) * 2024-03-12 2024-04-12 潍坊海之晨人工智能有限公司 一种运动型工业视觉实训平台

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109800699A (zh) * 2019-01-15 2019-05-24 珠海格力电器股份有限公司 图像识别方法、系统及装置
CN110334603A (zh) * 2019-06-06 2019-10-15 视联动力信息技术股份有限公司 身份验证系统
CN112862740B (zh) * 2019-11-28 2022-07-19 宁波微科光电股份有限公司 一种地铁障碍物检测方法
CN111339840B (zh) * 2020-02-10 2023-04-07 浙江大华技术股份有限公司 人脸检测方法和监控系统
CN111753757B (zh) * 2020-06-28 2021-06-18 浙江大华技术股份有限公司 一种图像识别处理方法及装置
CN112699843B (zh) * 2021-01-13 2024-09-06 上海云思智慧信息技术有限公司 一种身份识别方法及系统

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103336941A (zh) * 2006-07-19 2013-10-02 光谱辨识公司 多重生物测定多谱成像器
CN103914679A (zh) * 2013-01-05 2014-07-09 联想(北京)有限公司 一种图像识别方法、装置和电子设备
CN104599365A (zh) * 2014-12-31 2015-05-06 苏州福丰科技有限公司 基于三维人脸图像识别的平开式感应门控制系统及方法
CN106845449A (zh) * 2017-02-22 2017-06-13 浙江维尔科技有限公司 一种图像处理装置、方法以及人脸识别系统
US20170243054A1 (en) * 2014-12-11 2017-08-24 Lg Electronics Inc. Mobile terminal and control method thereof
CN107223258A (zh) * 2017-03-31 2017-09-29 中控智慧科技股份有限公司 图像采集方法和设备
CN107274517A (zh) * 2017-04-25 2017-10-20 合肥中科奔巴科技有限公司 融合三维人脸与掌静脉识别的智能门禁系统
CN107423714A (zh) * 2017-07-28 2017-12-01 中控智慧科技股份有限公司 图像获取设备和方法
CN108764091A (zh) * 2018-05-18 2018-11-06 北京市商汤科技开发有限公司 活体检测方法及装置、电子设备和存储介质
CN109800699A (zh) * 2019-01-15 2019-05-24 珠海格力电器股份有限公司 图像识别方法、系统及装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103136504B (zh) * 2011-11-28 2016-04-20 汉王科技股份有限公司 人脸识别方法及装置
CN103902961B (zh) * 2012-12-28 2017-02-15 汉王科技股份有限公司 一种人脸识别方法及装置
CN105718925A (zh) * 2016-04-14 2016-06-29 苏州优化智能科技有限公司 基于近红外和面部微表情的真人活体身份验证终端设备
CN105975908A (zh) * 2016-04-26 2016-09-28 汉柏科技有限公司 人脸识别方法及装置
CN107563304B (zh) * 2017-08-09 2020-10-16 Oppo广东移动通信有限公司 终端设备解锁方法及装置、终端设备
CN108090477A (zh) * 2018-01-23 2018-05-29 北京易智能科技有限公司 一种基于多光谱融合的人脸识别方法与装置
CN108776768A (zh) * 2018-04-19 2018-11-09 广州视源电子科技股份有限公司 图像识别方法及装置
CN108764058B (zh) * 2018-05-04 2021-05-25 吉林大学 一种基于热成像效应的双摄像头人脸活体检测方法

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103336941A (zh) * 2006-07-19 2013-10-02 光谱辨识公司 多重生物测定多谱成像器
CN103914679A (zh) * 2013-01-05 2014-07-09 联想(北京)有限公司 一种图像识别方法、装置和电子设备
US20170243054A1 (en) * 2014-12-11 2017-08-24 Lg Electronics Inc. Mobile terminal and control method thereof
CN104599365A (zh) * 2014-12-31 2015-05-06 苏州福丰科技有限公司 基于三维人脸图像识别的平开式感应门控制系统及方法
CN106845449A (zh) * 2017-02-22 2017-06-13 浙江维尔科技有限公司 一种图像处理装置、方法以及人脸识别系统
CN107223258A (zh) * 2017-03-31 2017-09-29 中控智慧科技股份有限公司 图像采集方法和设备
CN107274517A (zh) * 2017-04-25 2017-10-20 合肥中科奔巴科技有限公司 融合三维人脸与掌静脉识别的智能门禁系统
CN107423714A (zh) * 2017-07-28 2017-12-01 中控智慧科技股份有限公司 图像获取设备和方法
CN108764091A (zh) * 2018-05-18 2018-11-06 北京市商汤科技开发有限公司 活体检测方法及装置、电子设备和存储介质
CN109800699A (zh) * 2019-01-15 2019-05-24 珠海格力电器股份有限公司 图像识别方法、系统及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11295538B2 (en) 2020-02-20 2022-04-05 Asustek Computer Inc. Method and equipment for constructing three-dimensional face model
CN114926717A (zh) * 2022-04-18 2022-08-19 杭州海康威视数字技术股份有限公司 对象特征融合方法、装置以及存储介质
CN117876429A (zh) * 2024-03-12 2024-04-12 潍坊海之晨人工智能有限公司 一种运动型工业视觉实训平台
CN117876429B (zh) * 2024-03-12 2024-06-07 潍坊海之晨人工智能有限公司 一种运动型工业视觉实训系统

Also Published As

Publication number Publication date
CN109800699A (zh) 2019-05-24

Similar Documents

Publication Publication Date Title
WO2020147346A1 (zh) 图像识别方法、系统及装置
TWI759613B (zh) 基於多視角的多人臉場景刷臉支付風險防控方法和設備
CN107093171B (zh) 一种图像处理方法及装置、系统
CN107392958B (zh) 一种基于双目立体摄像机确定物体体积的方法及装置
CN109426785B (zh) 一种人体目标身份识别方法及装置
CN105912912B (zh) 一种终端用户身份登录方法和系统
WO2019071664A1 (zh) 结合深度信息的人脸识别方法、装置及存储介质
TW201405445A (zh) 基於雙攝影機之真實人臉識別系統及其方法
US20200257890A1 (en) Method and device for determining path of human target
CN106981078B (zh) 视线校正方法、装置、智能会议终端及存储介质
TW201627917A (zh) 人臉活體檢測方法和裝置
CN106844492A (zh) 一种人脸识别的方法、客户端、服务器及系统
CN101923641A (zh) 一种改进的人脸识别方法
CN111383256B (zh) 图像处理方法、电子设备及计算机可读存储介质
CN111680574A (zh) 一种人脸检测方法、装置、电子设备和存储介质
CN106295640A (zh) 一种智能终端的物体识别方法和装置
CN108764053A (zh) 图像处理方法、装置、计算机可读存储介质和电子设备
CN108764100B (zh) 一种目标行为检测方法及服务器
WO2017028674A1 (zh) 可视化远程控制可触控设备的方法、系统和相关设备
CN111383255A (zh) 图像处理方法、装置、电子设备及计算机可读存储介质
CN112818874B (zh) 一种图像处理方法、装置、设备及存储介质
CN112257617B (zh) 多模态目标识别方法和系统
CN107832598A (zh) 解锁控制方法及相关产品
CN109583296A (zh) 一种防止误检测方法、装置、系统及计算机存储介质
CN110800020B (zh) 一种图像信息获取方法、图像处理设备及计算机存储介质

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: 19910641

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19910641

Country of ref document: EP

Kind code of ref document: A1