WO2019090617A1 - 一种基于智能终端的人数统计方法及人数统计系统 - Google Patents

一种基于智能终端的人数统计方法及人数统计系统 Download PDF

Info

Publication number
WO2019090617A1
WO2019090617A1 PCT/CN2017/110246 CN2017110246W WO2019090617A1 WO 2019090617 A1 WO2019090617 A1 WO 2019090617A1 CN 2017110246 W CN2017110246 W CN 2017110246W WO 2019090617 A1 WO2019090617 A1 WO 2019090617A1
Authority
WO
WIPO (PCT)
Prior art keywords
face
real
image
module
face image
Prior art date
Application number
PCT/CN2017/110246
Other languages
English (en)
French (fr)
Inventor
詹昌松
Original Assignee
深圳传音通讯有限公司
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 深圳传音通讯有限公司 filed Critical 深圳传音通讯有限公司
Priority to PCT/CN2017/110246 priority Critical patent/WO2019090617A1/zh
Publication of WO2019090617A1 publication Critical patent/WO2019090617A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
    • G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity

Definitions

  • the present invention relates to the technical field of intelligent terminals, and in particular, to a method for counting people based on intelligent terminals and a system for counting people.
  • Smart terminals have become an increasingly indispensable part of today's life. With the rapid development of smart terminals, its built-in functions are becoming more diverse. Listening to music, business, games, and taking pictures has become a must-have feature for smart terminals. At present, in order to further improve the functions and user experience of smart terminals, technicians have continuously innovated and optimized existing functions.
  • the invention provides a method for counting people based on intelligent terminals and a population counting system. It does not require manual statistical operations, and at the same time ensures the correctness of statistical results, saves time and effort, and is easy to operate, providing users with a more convenient way to count people.
  • the invention provides a method for counting people based on intelligent terminals, and the method for counting people includes the following steps:
  • a statistical list containing the information of the unreceived objects is output.
  • the method further includes:
  • the face region is extracted to generate the real-to-face image displaying the face region.
  • the step of matching the real-to-face image with the face-to-face image that is preset to the object in the smart terminal further includes:
  • the statistical list includes one or more of the number of people, name, gender, age, unit, address, and contact information of the unrepresented person.
  • the method for counting people further comprises:
  • an unseen reminder is sent to the unreached object.
  • the present invention further provides a smart terminal-based population counting system, the number counting system comprising: a capture module, an acquisition module, a matching module, a marking module, a retrieval module, and an output module;
  • the capturing module captures a captured image of a face including at least one real object
  • the acquiring module is communicably connected to the capturing module and the matching module, and acquires a real-to-face image of each of the real-object objects;
  • the matching module is communicably connected to the acquiring module and the marking module, and matches the real-to-face image with a face-to-face image of the object that is preset to the smart terminal;
  • the marking module is in communication with the matching module and the calling module, and marks the unfinished image that fails to match the real-face image in the facial image;
  • the calling module is communicably connected to the marking module and the output module, and according to the unfinished image, the unobtained object information associated with the unfinished image is retrieved;
  • the output module is in communication with the earth module, and outputs a statistical list including the unreceived object information.
  • the obtaining module further comprises:
  • a recognition unit that identifies a face position of each of the real objects in the captured image
  • An extracting unit extracts the face region to generate the real-to-face image displaying the face region.
  • the matching module further comprises:
  • Reading unit reading the image to the face
  • a calculating unit calculating a similarity between the face feature and the real face feature in the real face image
  • the determining unit determines that the face image corresponding to the face feature corresponds to the face image
  • the real-to-face image corresponding to the facial feature is successfully matched.
  • the statistical list includes one or more of the number of people, name, gender, age, unit, address, and contact information of the unrepresented person.
  • the demographic system further includes a reminder module
  • the reminding module is in communication with the calling module
  • an unseen reminder is sent to the unreached object.
  • FIG. 1 is a schematic flowchart of a method for counting people based on an intelligent terminal according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of a method for counting people based on an intelligent terminal according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of a method for counting people based on an intelligent terminal according to an embodiment of the present invention
  • FIG. 4 is a schematic structural diagram of a population counting system based on an intelligent terminal according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a population counting system based on an intelligent terminal according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a population counting system based on an intelligent terminal according to an embodiment of the present invention.
  • the smart terminal can be implemented in various forms.
  • the terminal described in the present invention may include a mobile terminal such as a mobile phone, a smart phone, a notebook computer, a PDA (Personal Digital Assistant), a PAD (Tablet), a PMP (Portable Multimedia Player), a navigation device, and the like, and such as Fixed terminal for digital TV, desktop computer, etc.
  • a mobile terminal such as a mobile phone, a smart phone, a notebook computer, a PDA (Personal Digital Assistant), a PAD (Tablet), a PMP (Portable Multimedia Player), a navigation device, and the like
  • the terminal is a smart terminal.
  • those skilled in the art will appreciate that configurations in accordance with embodiments of the present invention can be applied to fixed type terminals in addition to components that are specifically for mobile purposes.
  • the method for counting people includes the following steps:
  • the method for counting people when the user needs to count the number of people present, it is necessary to first capture a captured image of the actual object that is actually present through the camera of the smart terminal (front, rear or external camera). Specifically, the captured image needs to include all the face regions of the real object, so that the subsequent smart terminal can recognize the real object.
  • the smart terminal After capturing the captured image of the face containing the real object, the smart terminal automatically recognizes the facial features of each real object in the captured image through the face recognition technology, and determines the facial contour to obtain each real
  • the face image of the object is the real to face image. According to the acquisition of the actual face image, the number of specific real objects and the number of real objects can be determined.
  • the pre-stored object in the smart terminal includes a name, a gender, a contact method, and a face image to match the real-face image to determine whether all the objects should be present.
  • the smart terminal marks such an object to be marked to indicate that the marked object to be actually is not the object, and is currently in an absent state that has not arrived at the scene.
  • the smart terminal can directly retrieve the unobtained object information associated with the unfinished face image according to the marked faceless image, such as name, gender, age, unit, address, contact information, and the like. This type of information can be stored in the smart terminal by pre-inputting, so that it can be retrieved at any time.
  • the statistical list includes one or more of the number of people, names, genders, ages, units, addresses, and contacts of the unreceived subject.
  • the smart terminal After the information of the object is not retrieved, the smart terminal outputs a statistics list of the number of people, and the statistics list includes the unobtained object information obtained by the statistic list, and the statistician can directly read the unreached object.
  • the number of people and personal information such as the name, gender, age, unit, address, and contact information of the unsuccessful person are convenient for the statistician to understand the simple information of the unreached object, and to facilitate the subsequent contact with the unreached object.
  • said method of counting people further comprises: issuing an unresolved reminder to said unreached object based on said unreceived object information.
  • the smart terminal when the smart terminal recognizes that the object has not arrived and acquires the object information according to the captured captured image, the smart terminal may further issue a reminder to the object according to the contact information in the object information to reach the reminder.
  • the effect of not arriving at the object as soon as possible is also achieved, and it is also possible to prevent the statistician from following the notification that the object has not arrived as soon as possible.
  • the step of acquiring a real-to-face image of each of the real-object objects further includes:
  • the smart terminal After capturing the captured image of the face including one or more real objects, the smart terminal recognizes the face of each of the captured images in the captured captured image with its built-in face recognition device, and extracts the real The face feature information of the object is obtained, and the face position of the real object is obtained. Through the face recognition device, the smart terminal can accurately know the number of people in the captured image, that is, obtain the actual number of people.
  • a face region capable of displaying the entire face contour is delineated for each real object according to the contour of the real object face.
  • the image in the face area is extracted from the background of the captured image as a real-to-face image of the real object.
  • the step of matching the real-to-face image with the face-to-face image that is preset to the object in the smart terminal further includes:
  • the smart terminal After the smart terminal extracts the real-to-face image from the captured image, in order to match the real-to-face image with the facial image stored in the smart terminal, the image corresponding to the facial image needs to be stored in the smart terminal.
  • the library issues a read permission request, and after obtaining the permission of the image library, the read image permission can be read before the image in the image library can be read.
  • the face recognition device built in the smart terminal After reading the image to the face, the face recognition device built in the smart terminal further performs preprocessing corresponding to the face image, including: binarization, grayscale, edge extraction, etc., to achieve corresponding to the object Extract to the facial features in the face image.
  • the smart terminal After obtaining the facial features, the smart terminal matches the facial features with the real facial features in the captured facial images, and determines the similarity between the two.
  • the similarity judgment is No match.
  • the real-to-face image extracted from the captured image obtained by the capture does not necessarily reach the frontal image of the user, and the captured real-to-face image may have a certain expression on the preset facial image. Deviation, therefore, the smart terminal determines that the matching between the facial feature and the real facial feature is greater than or equal to 90% based on the possible gap, that is, the smart terminal considers that the real object includes the Should go to the object.
  • the number of people counting system includes: a capturing module, an acquiring module, a matching module, a marking module, a calling module, and an output module;
  • the capture module capturing a captured image of the face comprising at least one real object
  • the camera in the capture module of the number counting system must first be used to actually arrive at the actual object. Capture a captured image. Specifically, the captured image needs to include all the face regions of the real object, so that the subsequent smart terminal can recognize the real object.
  • the acquisition module is communicatively coupled to the capture module and the matching module to obtain a real-to-face image of each of the real-to-object objects;
  • the acquisition module of the population counting system automatically recognizes the facial features of each real object in the captured image through the face recognition technology, and determines the facial contour to Get the face image of each real object, that is, the real to the face image. According to the acquisition of the actual face image, the number of specific real objects and the number of real objects can be determined.
  • the matching module is in communication with the acquiring module and the marking module, and matching the real-to-face image with a face-to-face image of the object to be preset preset in the smart terminal;
  • the matching module of the population statistics system matches the obtained real-to-face image with the facial image in the smart terminal to determine whether the image of the object should be up to the face.
  • the captured image obtained by the capture it is determined whether the object should have actually arrived.
  • the number of pre-existing objects in the number counting system includes the name, gender, contact information and facial image, so that the matching module can match the facial image with the real-to-face image to determine whether all the objects should be turn up.
  • the marking module is in communication with the matching module and the calling module, and marks the unfinished image that fails to match the real-face image in the facial image;
  • the marking module of the number counting system marks such an object to be marked to indicate that the marked object to be actually is not the object, and is currently in an absent state that has not arrived at the scene.
  • the call module is communicably connected to the mark module and the output module, and according to the unfinished image, the unobtained object information associated with the unfinished image is retrieved;
  • the retrieval module can directly retrieve the unobtained object information associated with the unfinished facial image according to the marked non-face image, such as name, gender, age, unit, address, contact information, and the like. This type of information can be stored in the smart terminal by pre-inputting, so that it can be retrieved at any time.
  • the output module is in communication with the earth module and outputs a statistical list containing the unreceived object information.
  • the statistical list includes one or more of the number of people, names, genders, ages, units, addresses, and contacts of the unreceived subject.
  • the output module of the population statistics system When the retrieval module retrieves the information to the object, the output module of the population statistics system outputs a population count column.
  • the table includes the unobtained information obtained in the statistics list. Through the statistical list, the statistician can directly read the number of unreached objects and the names, genders, ages, units, addresses, and contacts of the unreached objects. Personal information such as methods, it is convenient for statisticians to understand the simple information of the unreached object, and it is also convenient for subsequent contact with the unreached object.
  • the number counting system further includes a reminder module
  • the reminding module is in communication with the calling module; and according to the unreceived object information, an unseen reminder is sent to the unreached object.
  • the reminding module of the further number counting system may send the unreceived object according to the contact information in the object information.
  • a reminder in order to achieve the effect of reminding and urging the target not to arrive as soon as possible, at the same time, it can also prevent the statistician from following the notification that the object has not arrived at the scene as soon as possible.
  • the acquiring module further includes: an identifying unit that identifies a face position of each of the real objects in the captured image;
  • the recognition unit in the acquisition module combines its built-in face recognition device with each of the captured captured images.
  • the face is identified, and the facial feature information of the real object is extracted, and the face position of the real object is obtained.
  • the smart terminal can accurately know the number of people in the captured image, that is, obtain the actual number of people.
  • Delineating unit according to the position of the face, delineating a corresponding facial region for each of the real objects
  • the demarcation unit in the acquisition module delimits each real object according to the contour of the real object surface A face area capable of displaying the entire facial contour.
  • An extraction unit that extracts the face region to generate the real-to-face image that displays the face region.
  • the extraction unit of the acquisition module extracts the image in the face region from the background of the captured image as a real-to-face image of the real object.
  • FIG. 6 is a schematic structural diagram of a system for counting people based on an intelligent terminal according to an embodiment of the present invention.
  • the matching module further includes:
  • the reading unit in the matching module needs to store to the smart terminal.
  • a read permission request should be issued to the image library of the face image, and after the image library allows, the read permission is obtained to read the face image in the image library.
  • a processing unit that performs facial feature extraction on the applied object according to the image to be faced
  • the face recognition device built in the processing unit of the acquisition module further performs preprocessing corresponding to the facial image, including: binarization, grayscale, edge extraction, and the like.
  • the extraction is performed to achieve a face-to-face feature corresponding to the object in the face image.
  • a calculation unit that calculates a similarity of the face-to-face feature to the real-to-face feature in the real-to-face image
  • the computing unit matches the facial features with the real facial features in the captured partial images in the captured captured images to determine the similarity between the two. According to the similarity between the two, it is judged whether it matches.
  • the determining unit determines the face image corresponding to the face feature and the real The real-to-face image corresponding to the facial feature is successfully matched;
  • the real-to-face image extracted from the captured image captured by the capture module is not necessarily true to the user. Face image, and the actual face image captured by the capture module may have a certain deviation from the preset face image. Therefore, the judgment unit in the matching module determines the gap based on the possible existence when calculating When the unit calculation should reach the face feature and the real-to-face feature is greater than or equal to 90%, it can be determined that the matching is successful, that is, the population statistics system considers that the real object includes the corresponding object.
  • the user can capture the captured image including the face of the real object, and combine the face recognition technology to identify the unreached object, and the output includes the unseen A statistical list of object information.
  • the invention has convenient operation and can obtain statistical results in a short time, thereby eliminating the trouble of manual counting, and also avoiding the counting error generated by manual operation, and providing a more optimized technology for the user to quickly complete the population statistics. Program.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Image Analysis (AREA)

Abstract

本发明提供了一种基于智能终端的人数统计方法及人数统计系统。人数统计方法包括以下步骤：捕捉一包含至少一个实到对象的脸部的拍摄图像；获取每一实到对象的实到脸部图像；将实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；标记应到脸部图像中，与实到脸部图像匹配失败的未到脸部图像；根据未到脸部图像，调取与未到脸部图像关联的未到对象信息；输出一包含未到对象信息的统计列表。人数统计系统包括：捕捉模块、获取模块、匹配模块、标记模块、调取模块及输出模块。采用本发明的技术方案进行人数统计时，既无需人工统计操作，同时又能够确保统计结果的正确性，省时省力，操作便捷，为用户提供一种更为简便的人数统计方式。

Description

一种基于智能终端的人数统计方法及人数统计系统 技术领域
本发明涉及智能终端的技术领域，尤其涉及一种基于智能终端的人数统计方法及人数统计系统。
背景技术
智能终端成为了当今生活中越来越不可或缺的一部分。随着智能终端的飞速发展，其内置的功能也越来越多样化。听音乐、商务、游戏、拍照已然成为了智能终端必备的成熟功能。目前，为进一步改善智能终端的功能和用户体验，技术人员针对现有的功能作出了不断的革新和优化。
除了智能终端本身自带的基础功能以外，结合多种多样的应用程序，基于智能终端可以实现的功能已远远超乎了传统思维。传统上，报数或者点名为最直接的人数统计方法，但该方法的缺陷在于，需要花费大量的人工操作，并且当需要统计的人数过多时，往往会出现计算的偏差，且为此，往往统计人员需要重复统计多次以确保统计结果的正确性。
为此，需要提供一种基于智能终端的人数统计方法及人数统计系统，既无需人工统计操作，同时又能够确保统计结果的正确性，省时省力，操作便捷，为用户提供一种更为简便的人数统计方式。
发明内容
本发明提出的一种基于智能终端的人数统计方法及人数统计系统。既无需人工统计操作，同时又能够确保统计结果的正确性，省时省力，操作便捷，为用户提供一种更为简便的人数统计方式。
本发明提供了一种基于智能终端的人数统计方法，所述人数统计方法包括以下步骤：
捕捉一包含至少一个实到对象的脸部的拍摄图像；
获取每一所述实到对象的实到脸部图像；
将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；
标记所述应到脸部图像中，与所述实到脸部图像匹配失败的未到脸部图像；
根据所述未到脸部图像，调取与所述未到脸部图像关联的未到对象信息；
输出一包含所述未到对象信息的统计列表。
优选地，获取每一所述实到对象的实到脸部图像的步骤中，进一步包括：
识别所述拍摄图像中的每一所述实到对象的脸部位置；
根据所述脸部位置，对每一所述实到对象划定一相应的脸部区域；
提取所述脸部区域，以生成显示所述脸部区域的所述实到脸部图像。
优选地，将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配的步骤中，进一步包括：
读取所述应到脸部图像；
根据所述应到脸部图像进行对所述应到对象的应到脸部特征提取；
计算所述应到脸部特征与所述实到脸部图像中的实到脸部特征的相似度；
当所述应到脸部特征与所述实到脸部特征的相似度大于等于90％时，则判断所述应到脸部特征对应的所述应到脸部图像与所述实到脸部特征对应的所述实到脸部图像匹配成功。
优选地，所述统计列表包括所述未到对象的人数、姓名、性别、年龄、单位、地址、联系方式中的一种或多种。
优选地，所述人数统计方法进一步包括：
根据所述未到对象信息，向所述未到对象发出一未到提醒。
本发明进一步提供了一种基于智能终端的人数统计系统，所述人数统计系统包括：捕捉模块、获取模块、匹配模块、标记模块、调取模块及输出模块；
所述捕捉模块，捕捉一包含至少一个实到对象的脸部的拍摄图像；
所述获取模块，与所述捕捉模、匹配模块通讯连接，获取每一所述实到对象的实到脸部图像；
所述匹配模块，与所述获取模块、标记模块通讯连接，将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；
所述标记模块，与所述匹配模块、调取模块通讯连接，标记所述应到脸部图像中，与所述实到脸部图像匹配失败的未到脸部图像；
所述调取模块，与所述标记模块、输出模块通讯连接，根据所述未到脸部图像，调取与所述未到脸部图像关联的未到对象信息；
所述输出模块，与所述地球模块通讯连接，输出一包含所述未到对象信息的统计列表。
优选地，所述获取模块进一步包括：
识别单元，识别所述拍摄图像中的每一所述实到对象的脸部位置；
划定单元，根据所述脸部位置，对每一所述实到对象划定一相应的脸部区域；
提取单元，提取所述脸部区域，以生成显示所述脸部区域的所述实到脸部图像。
优选地，所述匹配模块进一步包括：
读取单元，读取所述应到脸部图像；
处理单元，根据所述应到脸部图像进行对所述应到对象的应到脸部特征提取；
计算单元，计算所述应到脸部特征与所述实到脸部图像中的实到脸部特征的相似度；
当所述应到脸部特征与所述实到脸部特征的相似度大于等于90％时，则判断单元判断所述应到脸部特征对应的所述应到脸部图像与所述实到脸部特征对应的所述实到脸部图像匹配成功。
优选地，所述统计列表包括所述未到对象的人数、姓名、性别、年龄、单位、地址、联系方式中的一种或多种。
优选地，所述人数统计系统进一步包括一提醒模块；
所述提醒模块，与所述调取模块通讯连接；
根据所述未到对象信息，向所述未到对象发出一未到提醒。
与现有技术相比较，本发明的技术优势在于：
1.基于拍照和人脸识别技术，快速辨识未到对象；
2.无需人工统计操作；
3.统计结果正确性高；
4.省时省力；
5.操作简便。
附图说明
图1为符合本发明实施例中的一种基于智能终端的人数统计方法的流程示意图；
图2为符合本发明实施例中的一种基于智能终端的人数统计方法的流程示意图；
图3为符合本发明实施例中的一种基于智能终端的人数统计方法的流程示意图；
图4为符合本发明实施例中的一种基于智能终端的人数统计系统的结构示意图；
图5为符合本发明实施例中的一种基于智能终端的人数统计系统的结构示意图；
图6为符合本发明实施例中的一种基于智能终端的人数统计系统的结构示意图。
具体实施方式
以下结合附图及具体实施例详细阐述本发明的优异性。
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本公开相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本公开的一些方面相一致的装置和方法的例子。
在本公开使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本公开。在本公开和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。
智能终端可以以各种形式来实施。例如，本发明中描述的终端可以包括诸如移动电话、智能电话、笔记本电脑、PDA(个人数字助理)、PAD(平板电脑)、PMP(便携式多媒体播放器)、导航装置等等的移动终端以及诸如数字TV、台式计算机等等的固定终端。下面，假设终端是智能终端。然而，本领域技术人员将理解的是，除了特别用于移动目的的元件之外，根据本发明的实施方式的构造也能够应用于固定类型的终端。
在后续的描述中，使用用于表示元件的诸如“模块”的后缀仅为了有利于本发明的说明，其本身并没有特定的意义。
参阅图1，为符合本发明实施例中的一种基于智能终端的人数统计方法的流程示意图。如图所示，本实施例中，所述人数统计方法包括以下步骤：
-捕捉一包含至少一个实到对象的脸部的拍摄图像；
基于本发明提供的人数统计方法，当用户需要对到场人数进行统计时，需要首先通过智能终端的摄像头(前置、后置或外接摄像头皆可)对实际到场的实到对象捕捉一拍摄图像。具体地，该拍摄图像需要包含所有实到对象的脸部区域，以便后续智能终端对实到对象的识别。
-获取每一所述实到对象的实到脸部图像；
捕捉到包含实到对象的脸部的拍摄图像后，智能终端自动通过脸部识别技术对拍摄图像中每一实到对象的脸部特征进行识别，确定脸部轮廓，以获取到每一实到对象的脸部图像，即为实到脸部图像。根据实到脸部图像的获取，可以确定具体的实到对象以及实到对象的人数。
-将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；
获取到实到脸部图像后，将所得实到脸部图像与智能终端中的应到脸部图像进行匹配，以判断应到对象的应到脸部图像是否都存在捕捉所得的拍摄图像中，即判断应到对象是否都已实际到场。其中，智能终端中预先存有应到对象包括名字、性别、联系方式及脸部图像，以便与实到脸部图像的进行匹配，判断所有应到对象是否到场。
-标记所述应到脸部图像中，与所述实到脸部图像匹配失败的未到脸部图像；
当应到脸部图像中存在一个或多个图像未能与实到脸部图像匹配成功，即存在一个或多个应到对象当前处于缺席状态。判断完成后，智能终端将此类应到对象进行标记，以表示该被标记的应到对象实际为未到对象，当前处于一未抵达现场的缺席状态。
-根据所述未到脸部图像，调取与所述未到脸部图像关联的未到对象信息；
智能终端根据被标记的未到脸部图像，可以直接调取到与该未到脸部图像关联设置的未到对象信息，例如：姓名、性别、年龄、单位、地址、联系方式等。该类信息可以通过预先输入储存于智能终端中，以备随时调取。
-输出一包含所述未到对象信息的统计列表。
-优选地，所述统计列表包括所述未到对象的人数、姓名、性别、年龄、单位、地址、联系方式中的一种或多种。
当调取到未到对象信息后，智能终端会输出一人数统计列表，在该统计列表中包括了调取所得的未到对象信息，通过该统计列表，统计人员可以直接读取未到对象的人数以及未到对象的姓名、性别、年龄、单位、地址、联系方式等个人信息，便于统计人员了解未到对象的简单信息，也便于后续与该未到对象取得联系。
-优选地，所述人数统计方法进一步包括：根据所述未到对象信息，向所述未到对象发出一未到提醒。
优选地，当智能终端根据捕捉到的拍摄图像识别出未到对象并获取到未到对象信息后，可进一步根据未到对象信息中的联系方式，向未到对象发出一提醒，以达到提醒督促未到对象尽快到场的效果，同时也能避免统计人员后续一个个陆续通知未到对象尽快到场的通知工作。
参阅图2，为符合本发明实施例中的一种基于智能终端的人数统计方法的流程示意图。如图所示，在一优选实施例中，获取每一所述实到对象的实到脸部图像的步骤中，进一步包括：
-识别所述拍摄图像中的每一所述实到对象的脸部位置；
当捕捉到包含一个或多个实到对象的脸部的拍摄图像后，智能终端结合其内置的人脸识别装置对捕捉所得的拍摄图像中的每一实到对象的脸部进行识别，提取实到对象的脸部特征信息，获取实到对象的脸部位置。通过人脸识别装置，智能终端可以精确获知拍摄图像中的人数，即获取实到人数。
-根据所述脸部位置，对每一所述实到对象划定一相应的脸部区域；
当通过人脸识别获取到拍摄图像中的实到对象的脸部位置后，根据实到对象脸部的轮廓，为每一实到对象划定一能够显示整个脸部轮廓的脸部区域。
-提取所述脸部区域，以生成显示所述脸部区域的所述实到脸部图像。
划定脸部区域后，从拍摄图像的背景中提取脸部区域中的图像，以作为实到对象的实到脸部图像。
参阅图3，为符合本发明实施例中的一种基于智能终端的人数统计方法的流程示意图。如图所示，在一优选实施例中，将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配的步骤中，进一步包括：
-读取所述应到脸部图像；
当智能终端从拍摄图像中提取到实到脸部图像后，为将实到脸部图像与智能终端中存有的应到脸部图像进行匹配，需向智能终端储存应到脸部图像的图片库发出一读取权限请求，经图片库允许后，获取到读取权限后方可读取图片库中的应到脸部图像。
-根据所述应到脸部图像进行对所述应到对象的应到脸部特征提取；
读取到应到脸部图像后，智能终端内置的人脸识别装置进一步对应到脸部图像进行预处理，包括：二值化、灰度化、边缘提取等操作，以实现对应到对象的应到脸部图像中的应到脸部特征进行提取。
-计算所述应到脸部特征与所述实到脸部图像中的实到脸部特征的相似度；
当获得应到脸部特征后，智能终端将应到脸部特征与捕捉所得的拍摄图像中的实到脸部图像中的实到脸部特征进行匹配，判断二者的相似度，以根据二者的相似度判断是 否匹配。
-当所述应到脸部特征与所述实到脸部特征的相似度大于等于90％时，则判断所述应到脸部特征对应的所述应到脸部图像与所述实到脸部特征对应的所述实到脸部图像匹配成功；
基于捕捉所得的拍摄图像中提取到的实到脸部图像不一定实到用户的正面脸部图像，并且捕捉所得的实到脸部图像可能与预设的应到脸部图像存在一定表情上的偏差，因此，智能终端基于可能存有的差距判定当应到脸部特征与实到脸部特征大于等于90％时，即可判定二者匹配成功，即智能终端认为该实到对象中包含该应到对象。
参阅图4，为符合本发明实施例中的一种基于智能终端的人数统计系统的结构示意图。如图所示，本实施例中，所述人数统计系统包括：捕捉模块、获取模块、匹配模块、标记模块、调取模块及输出模块；
-所述捕捉模块，捕捉一包含至少一个实到对象的脸部的拍摄图像；
基于本发明提供的人数统计系统，当用户需要对到场人数进行统计时，需要首先通过人数统计系统的捕捉模块中的摄像头(前置、后置或外接摄像头皆可)对实际到场的实到对象捕捉一拍摄图像。具体地，该拍摄图像需要包含所有实到对象的脸部区域，以便后续智能终端对实到对象的识别。
-所述获取模块，与所述捕捉模、匹配模块通讯连接，获取每一所述实到对象的实到脸部图像；
捕捉模块捕捉到包含实到对象的脸部的拍摄图像后，人数统计系统的获取模块自动通过脸部识别技术对拍摄图像中每一实到对象的脸部特征进行识别，确定脸部轮廓，以获取到每一实到对象的脸部图像，即为实到脸部图像。根据实到脸部图像的获取，可以确定具体的实到对象以及实到对象的人数。
-所述匹配模块，与所述获取模块、标记模块通讯连接，将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；
获取模块获取到实到脸部图像后，人数统计系统的匹配模块将所得实到脸部图像与智能终端中的应到脸部图像进行匹配，以判断应到对象的应到脸部图像是否都存在捕捉所得的拍摄图像中，即判断应到对象是否都已实际到场。其中，人数统计系统中预先存有应到对象包括名字、性别、联系方式及脸部图像，以便匹配模块能够将应到脸部图像与实到脸部图像的进行匹配，判断所有应到对象是否到场。
-所述标记模块，与所述匹配模块、调取模块通讯连接，标记所述应到脸部图像中，与所述实到脸部图像匹配失败的未到脸部图像；
通过匹配模块匹配，当应到脸部图像中存在一个或多个图像未能与实到脸部图像匹配成功，即存在一个或多个应到对象当前处于缺席状态。匹配模块匹配完成后，人数统计系统的标记模块将此类应到对象进行标记，以表示该被标记的应到对象实际为未到对象，当前处于一未抵达现场的缺席状态。
-所述调取模块，与所述标记模块、输出模块通讯连接，根据所述未到脸部图像，调取与所述未到脸部图像关联的未到对象信息；
调取模块根据被标记的未到脸部图像，可以直接调取到与该未到脸部图像关联设置的未到对象信息，例如：姓名、性别、年龄、单位、地址、联系方式等。该类信息可以通过预先输入储存于智能终端中，以备随时调取。
-所述输出模块，与所述地球模块通讯连接，输出一包含所述未到对象信息的统计列表。
-优选地，所述统计列表包括所述未到对象的人数、姓名、性别、年龄、单位、地址、联系方式中的一种或多种。
当调取模块调取到未到对象信息后，人数统计系统的输出模块会输出一人数统计列 表，在该统计列表中包括了调取所得的未到对象信息，通过该统计列表，统计人员可以直接读取未到对象的人数以及未到对象的姓名、性别、年龄、单位、地址、联系方式等个人信息，便于统计人员了解未到对象的简单信息，也便于后续与该未到对象取得联系。
-优选地，
所述人数统计系统进一步包括一提醒模块；
所述提醒模块，与所述调取模块通讯连接；根据所述未到对象信息，向所述未到对象发出一未到提醒。
优选地，当调取模块根据捕捉到的拍摄图像识别出未到对象并获取到未到对象信息后，可进一步人数统计系统的提醒模块根据未到对象信息中的联系方式，向未到对象发出一提醒，以达到提醒督促未到对象尽快到场的效果，同时也能避免统计人员后续一个个陆续通知未到对象尽快到场的通知工作。
参阅图5，为符合本发明实施例中的一种基于智能终端的人数统计系统的结构示意图。如图所示，在一优选实施例中，所述获取模块进一步包括：-识别单元，识别所述拍摄图像中的每一所述实到对象的脸部位置；
当捕捉模块捕捉到包含一个或多个实到对象的脸部的拍摄图像后，获取模块中的识别单元会结合其内置的人脸识别装置对捕捉所得的拍摄图像中的每一实到对象的脸部进行识别，提取实到对象的脸部特征信息，获取实到对象的脸部位置。通过人脸识别装置，智能终端可以精确获知拍摄图像中的人数，即获取实到人数。
-划定单元，根据所述脸部位置，对每一所述实到对象划定一相应的脸部区域；
当获取模块中的识别单元通过人脸识别获取到拍摄图像中的实到对象的脸部位置后，获取模块中的划定单元根据实到对象脸部的轮廓，为每一实到对象划定一能够显示整个脸部轮廓的脸部区域。
-提取单元，提取所述脸部区域，以生成显示所述脸部区域的所述实到脸部图像。
获取模块中的划定单元划定脸部区域后，获取模块的提取单元从拍摄图像的背景中提取脸部区域中的图像，以作为实到对象的实到脸部图像。
参阅图6，为符合本发明实施例中的一种基于智能终端的人数统计系统的结构示意图。如图所示，在一优选实施例中，所述匹配模块进一步包括：
-读取单元，读取所述应到脸部图像；
当获取模块从拍摄图像中提取到实到脸部图像后，为将实到脸部图像与智能终端中存有的应到脸部图像进行匹配，匹配模块中的读取单元需向智能终端储存应到脸部图像的图片库发出一读取权限请求，经图片库允许后，获取到读取权限后，以读取图片库中的应到脸部图像。
-处理单元，根据所述应到脸部图像进行对所述应到对象的应到脸部特征提取；
读取单元读取到应到脸部图像后，获取模块的处理单元中内置的人脸识别装置进一步对应到脸部图像进行预处理，包括：二值化、灰度化、边缘提取等操作，以实现对应到对象的应到脸部图像中的应到脸部特征进行提取。
-计算单元，计算所述应到脸部特征与所述实到脸部图像中的实到脸部特征的相似度；
当处理单元获得应到脸部特征后，计算单元将应到脸部特征与捕捉所得的拍摄图像中的实到脸部图像中的实到脸部特征进行匹配，判断二者的相似度，以根据二者的相似度判断是否匹配。
-当所述应到脸部特征与所述实到脸部特征的相似度大于等于90％时，则判断单元判断所述应到脸部特征对应的所述应到脸部图像与所述实到脸部特征对应的所述实到脸部图像匹配成功；
基于捕捉模块捕捉所得的拍摄图像中提取到的实到脸部图像不一定实到用户的正 面脸部图像，并且捕捉模块捕捉所得的实到脸部图像可能与预设的应到脸部图像存在一定表情上的偏差，因此，匹配模块中的判断单元基于可能存有的差距判定当计算单元计算应到脸部特征与实到脸部特征大于等于90％时，即可判定二者匹配成功，即人数统计系统认为该实到对象中包含该应到对象。
采用本发明提供的一种基于智能终端的人数统计方法及人数统计系统，用户可通过捕捉包含实到对象脸部的拍摄图像，结合人脸识别技术，识别出未到对象，以输出包含未到对象信息的统计列表。本发明操作便捷，能够在短时间内获得统计结果，既免去了人工计数的麻烦，同时也能够避免人工操作所产生的计数误差，为用户快速完成人数统计提供了一种更为优化的技术方案。
应当注意的是，本发明的实施例有较佳的实施性，且并非对本发明作任何形式的限制，任何熟悉该领域的技术人员可能利用上述揭示的技术内容变更或修饰为等同的有效实施例，但凡未脱离本发明技术方案的内容，依据本发明的技术实质对以上实施例所作的任何修改或等同变化及修饰，均仍属于本发明技术方案的范围内。

Claims (10)

  1. 一种基于智能终端的人数统计方法，其特征在于，
    所述人数统计方法包括以下步骤：
    捕捉一包含至少一个实到对象的脸部的拍摄图像；
    获取每一所述实到对象的实到脸部图像；
    将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；
    标记所述应到脸部图像中，与所述实到脸部图像匹配失败的未到脸部图像；
    根据所述未到脸部图像，调取与所述未到脸部图像关联的未到对象信息；
    输出一包含所述未到对象信息的统计列表。
  2. 如权利要求1所述的人数统计方法，其特征在于，
    获取每一所述实到对象的实到脸部图像的步骤中，进一步包括：
    识别所述拍摄图像中的每一所述实到对象的脸部位置；
    根据所述脸部位置，对每一所述实到对象划定一相应的脸部区域；
    提取所述脸部区域，以生成显示所述脸部区域的所述实到脸部图像。
  3. 如权利要求1所述的人数统计方法，其特征在于，
    将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配的步骤中，进一步包括：
    读取所述应到脸部图像；
    根据所述应到脸部图像进行对所述应到对象的应到脸部特征提取；
    计算所述应到脸部特征与所述实到脸部图像中的实到脸部特征的相似度；
    当所述应到脸部特征与所述实到脸部特征的相似度大于等于90％时，则判断所述应到脸部特征对应的所述应到脸部图像与所述实到脸部特征对应的所述实到脸部图像匹配成功。
  4. 如权利要求1所述的人数统计方法，其特征在于，
    所述统计列表包括所述未到对象的人数、姓名、性别、年龄、单位、地址、联系方式中的一种或多种。
  5. 如权利要求1-4任一项所述的人数统计方法，其特征在于，
    所述人数统计方法进一步包括：
    根据所述未到对象信息，向所述未到对象发出一未到提醒。
  6. 一种基于智能终端的人数统计系统，其特征在于，
    所述人数统计系统包括：捕捉模块、获取模块、匹配模块、标记模块、调取模块及输出模块；
    所述捕捉模块，捕捉一包含至少一个实到对象的脸部的拍摄图像；
    所述获取模块，与所述捕捉模、匹配模块通讯连接，获取每一所述实到对象的实到脸部图像；
    所述匹配模块，与所述获取模块、标记模块通讯连接，将所述实到脸部图像与所述智能终端中预设的应到对象的应到脸部图像进行匹配；
    所述标记模块，与所述匹配模块、调取模块通讯连接，标记所述应到脸部图像中，与所述实到脸部图像匹配失败的未到脸部图像；
    所述调取模块，与所述标记模块、输出模块通讯连接，根据所述未到脸部图像，调取与所述未到脸部图像关联的未到对象信息；
    所述输出模块，与所述地球模块通讯连接，输出一包含所述未到对象信息的统计列表。
  7. 如权利要求6所述的人数统计系统，其特征在于，
    所述获取模块进一步包括：
    识别单元，识别所述拍摄图像中的每一所述实到对象的脸部位置；
    划定单元，根据所述脸部位置，对每一所述实到对象划定一相应的脸部区域；
    提取单元，提取所述脸部区域，以生成显示所述脸部区域的所述实到脸部图像。
  8. 如权利要求6所述的人数统计系统，其特征在于，
    所述匹配模块进一步包括：
    读取单元，读取所述应到脸部图像；
    处理单元，根据所述应到脸部图像进行对所述应到对象的应到脸部特征提取；
    计算单元，计算所述应到脸部特征与所述实到脸部图像中的实到脸部特征的相似度；
    当所述应到脸部特征与所述实到脸部特征的相似度大于等于90％时，则判断单元判断所述应到脸部特征对应的所述应到脸部图像与所述实到脸部特征对应的所述实到脸部图像匹配成功。
  9. 如权利要求6所述的人数统计系统，其特征在于，
    所述统计列表包括所述未到对象的人数、姓名、性别、年龄、单位、地址、联系方式中的一种或多种。
  10. 如权利要求6-9任一项所述的人数统计系统，其特征在于，
    所述人数统计系统进一步包括一提醒模块；
    所述提醒模块，与所述调取模块通讯连接；
    根据所述未到对象信息，向所述未到对象发出一未到提醒。
PCT/CN2017/110246 2017-11-09 2017-11-09 一种基于智能终端的人数统计方法及人数统计系统 WO2019090617A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/110246 WO2019090617A1 (zh) 2017-11-09 2017-11-09 一种基于智能终端的人数统计方法及人数统计系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/110246 WO2019090617A1 (zh) 2017-11-09 2017-11-09 一种基于智能终端的人数统计方法及人数统计系统

Publications (1)

Publication Number Publication Date
WO2019090617A1 true WO2019090617A1 (zh) 2019-05-16

Family

ID=66438730

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/110246 WO2019090617A1 (zh) 2017-11-09 2017-11-09 一种基于智能终端的人数统计方法及人数统计系统

Country Status (1)

Country Link
WO (1) WO2019090617A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103345719A (zh) * 2013-07-03 2013-10-09 青岛大学 一种实验教学成绩评定系统和方法
CN103455958A (zh) * 2013-09-17 2013-12-18 云南大学 一种基于手机平台的课堂考勤方法
CN104183029A (zh) * 2014-09-02 2014-12-03 济南大学 一种便携式快速人群考勤方法
CN105303632A (zh) * 2015-10-22 2016-02-03 苏州工业园区服务外包职业学院 一种移动监控签到系统及工作方法
WO2016208120A1 (ja) * 2015-06-24 2016-12-29 パナソニックＩｐマネジメント株式会社 入退管理装置および入退管理方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103345719A (zh) * 2013-07-03 2013-10-09 青岛大学 一种实验教学成绩评定系统和方法
CN103455958A (zh) * 2013-09-17 2013-12-18 云南大学 一种基于手机平台的课堂考勤方法
CN104183029A (zh) * 2014-09-02 2014-12-03 济南大学 一种便携式快速人群考勤方法
WO2016208120A1 (ja) * 2015-06-24 2016-12-29 パナソニックＩｐマネジメント株式会社 入退管理装置および入退管理方法
CN105303632A (zh) * 2015-10-22 2016-02-03 苏州工业园区服务外包职业学院 一种移动监控签到系统及工作方法

Similar Documents

Publication Publication Date Title
US11074436B1 (en) Method and apparatus for face recognition
US9953216B2 (en) Systems and methods for performing actions in response to user gestures in captured images
US20150332439A1 (en) Methods and devices for hiding privacy information
WO2019071664A1 (zh) 结合深度信息的人脸识别方法、装置及存储介质
WO2020156034A1 (zh) 指纹录入方法及相关装置
RU2664003C2 (ru) Способ и устройство для определения ассоциированного пользователя
US10216976B2 (en) Method, device and medium for fingerprint identification
Vazquez-Fernandez et al. Built-in face recognition for smart photo sharing in mobile devices
JP2021034003A (ja) 人物識別方法、装置、電子デバイス、記憶媒体、及びプログラム
WO2017197778A1 (zh) 图像发送方法和装置
US9779294B2 (en) Methods and devices for classifying pictures
CN110941992B (zh) 微笑表情检测方法、装置、计算机设备及存储介质
CN111709414A (zh) Ar设备及其文字识别方法、装置和计算机可读存储介质
US20230316529A1 (en) Image processing method and apparatus, device and storage medium
WO2020078215A1 (zh) 基于视频的信息获取方法和装置
JPWO2017170384A1 (ja) 生体データ処理装置、生体データ処理システム、生体データ処理方法、生体データ処理プログラム、生体データ処理プログラムを記憶する記憶媒体
WO2019062347A1 (zh) 人脸识别方法及相关产品
CN106033539A (zh) 一种基于视频人脸识别的会议引导方法及系统
WO2013004037A1 (zh) 判断联系人列表中是否存在欲添加名片的方法
CN111797746B (zh) 人脸识别方法、装置及计算机可读存储介质
WO2022077139A1 (zh) 人脸识别方法、装置及可读存储介质
CN105426904A (zh) 照片处理方法、装置和设备
TWM457241U (zh) 結合擴增實境的圖像文字辨識系統
US10593077B2 (en) Associating digital ink markups with annotated content
US20170171462A1 (en) Image Collection Method, Information Push Method and Electronic Device, and Mobile Phone

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: 17931708

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: 17931708

Country of ref document: EP

Kind code of ref document: A1