WO2017000352A1 - 一种指纹识别方法、装置和终端 - Google Patents

一种指纹识别方法、装置和终端 Download PDF

Info

Publication number
WO2017000352A1
WO2017000352A1 PCT/CN2015/085843 CN2015085843W WO2017000352A1 WO 2017000352 A1 WO2017000352 A1 WO 2017000352A1 CN 2015085843 W CN2015085843 W CN 2015085843W WO 2017000352 A1 WO2017000352 A1 WO 2017000352A1
Authority
WO
WIPO (PCT)
Prior art keywords
template
fingerprint
matching
active
feature value
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/CN2015/085843
Other languages
English (en)
French (fr)
Inventor
张明臣
成荣峰
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.)
Yulong Computer Telecommunication Scientific Shenzhen Co Ltd
Original Assignee
Yulong Computer Telecommunication Scientific Shenzhen Co Ltd
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 Yulong Computer Telecommunication Scientific Shenzhen Co Ltd filed Critical Yulong Computer Telecommunication Scientific Shenzhen Co Ltd
Publication of WO2017000352A1 publication Critical patent/WO2017000352A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
    • A61B5/0082—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes
    • A61B5/0088—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes for oral or dental tissue
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1365—Matching; Classification
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor

Definitions

  • the present invention relates to the field of fingerprint recognition technologies, and in particular, to a fingerprint identification method, apparatus, and terminal.
  • the process of fingerprint recognition is as follows: firstly, the fingerprint template is used to collect the fingerprint template; then the fingerprint template is initially processed and stored; then the fingerprint is collected and compared with the fingerprint template; and finally the comparison result is output.
  • the key technology of fingerprint recognition is the process of collecting fingerprints and comparing with templates, and it is rare for only one template in the field of high security and high precision.
  • the usual scene is: a fingerprint matches a large number of templates, such as attendance, access control, public security, and so on. Due to the large number of templates in the template library, the process of comparison is time consuming and difficult to accept. Therefore, there is a need for a fingerprint identification method, apparatus, and terminal to solve the above-described technical problems existing in the prior art.
  • the technical problem to be solved by the present invention is to provide a fingerprint identification method, device and terminal, which improve the speed of fingerprint recognition.
  • a fingerprint identification method comprising: a fingerprint collection step: acquiring a fingerprint image, extracting a fingerprint feature value from the fingerprint image; and selecting a template: selecting an active template from the template database, and extracting features of the active template a value, wherein the fingerprint template is marked with a matching success number, the active template is the highest number of matching successes in the fingerprint template; and the matching step: comparing the fingerprint feature value with the feature value of the active template ; if the match fails, Returning to the execution of the template selection step, if the matching is successful, the number of matching successes of the active template tag is increased once.
  • the selecting an active template from the template database includes: selecting a plurality of fingerprint templates from the template database according to the number of matching success times of the fingerprint template marking to form a template. a collection; an active template is selected from the set of templates.
  • the method further comprises: a reading step of: reading a new fingerprint template from the template database under the condition that the matching fails in the matching step, and adding the new fingerprint template to the template After the new fingerprint template is added to the template collection, an active template is selected from the template collection.
  • the number of matching success times of the new fingerprint template tag is the highest among the remaining fingerprint templates in the template database.
  • the fingerprint collecting step further includes: extracting a salient feature value from the fingerprint image in the process of acquiring a fingerprint image; and the matching step comprises: selecting the salient feature value and the feature value of the active template If the matching degree does not exceed the set threshold, the feature value of the next active template selected in the template selection step is compared; if the matching degree exceeds the set threshold, the fingerprint image is obtained after the fingerprint acquisition step is extracted. All the fingerprint feature values are compared with the feature values of the active template. If the matching fails, the template selection step is performed. If the matching is successful, the number of matching success times of the active template tag is increased once.
  • the invention also provides a device for fingerprint identification, comprising: a fingerprint collection module: for acquiring a fingerprint image to extract a fingerprint feature value from the fingerprint image; and a template selection module: for selecting an active template from the template database And the feature value of the active template is extracted, wherein the fingerprint template is marked with a matching success number, the active template is the highest number of matching successes of the marking in the fingerprint template; and the matching module is configured to use the fingerprint feature value If the matching fails, the template selection module selects the active template again to perform the next matching. If the matching succeeds, the matching success number of the active template label is increased once.
  • the template selection module is further configured to: sequentially select a plurality of fingerprint templates from the template database according to the number of matching success times of the fingerprint template mark to form a template set; select from the template set An active template.
  • the device further comprises a reading module, configured to: match the failed matching bar in the matching module Reading a new fingerprint template from the template database, and adding the new fingerprint template to the template collection; after adding the new fingerprint template to the template collection, the template selection module is An active template is selected from the template collection.
  • a reading module configured to: match the failed matching bar in the matching module Reading a new fingerprint template from the template database, and adding the new fingerprint template to the template collection; after adding the new fingerprint template to the template collection, the template selection module is An active template is selected from the template collection.
  • the present invention also provides a terminal comprising a communication bus, an input device, a memory, and a processor, wherein:
  • the communication bus is configured to implement connection communication between the input device, the memory, and the processor
  • the input device is configured to acquire a fingerprint image
  • the program stores a set of program codes, and the terminal calls the program code stored in the memory to perform the following operations:
  • the input device acquires a fingerprint image, and extracts a fingerprint feature value from the fingerprint image
  • the processor selects an active template from the template database, and extracts a feature value of the active template, where the fingerprint template is marked with a matching success number, and the active template is the highest matching success rate of the tag in the fingerprint template. ;
  • the processor compares the fingerprint feature value with a feature value of the active template
  • the processor adds the matching success number of the active template tag once.
  • the processor selects an active template from the template database, including:
  • An active template is selected from the set of templates.
  • the method further comprises:
  • the processor reads a new fingerprint template from the template database and adds the new fingerprint template to the template set under the condition that the matching fails;
  • the processor After the processor adds the new fingerprint template to the template set, the processor selects an active template from the template set.
  • the number of matching success times of the new fingerprint template tag is the highest among the remaining fingerprint templates of the template database.
  • the method further comprises:
  • the processor compares the fingerprint feature value with the feature value of the active template, including:
  • the feature value of the next active template selected in the template selection step is compared;
  • the matching degree exceeds the set threshold, compare all the fingerprint feature values extracted after acquiring the fingerprint image in the fingerprint collecting step with the feature values of the active template, and if the matching fails, return to execute the template selecting step, if If the match is successful, the number of matching successes of the active template tag is increased once.
  • the present invention has at least the following effects:
  • the fingerprint identification method of the invention improves the speed and efficiency of fingerprint recognition, and reduces the time that the user waits while performing fingerprint recognition.
  • FIG. 1 is a flowchart of a fingerprint identification method according to a first embodiment of the present invention
  • FIG. 2 is a flowchart of a fingerprint identification method according to a second embodiment of the present invention.
  • FIG. 3 is a flowchart of a fingerprint identification method according to a third embodiment of the present invention.
  • FIG. 4 is a flowchart of a fingerprint identification method according to a fourth embodiment of the present invention.
  • Figure 5 is a block diagram of a fingerprint recognition apparatus according to a fifth embodiment of the present invention.
  • Figure 6 is a block diagram of a fingerprint identification apparatus according to a sixth embodiment of the present invention.
  • Figure 7 is a block diagram of a fingerprint recognition apparatus according to a seventh embodiment of the present invention.
  • Figure 8 is a block diagram of a terminal of an eighth embodiment of the present invention.
  • the fingerprint identification method provided by the present invention may The fingerprint recognition is completed faster, and the work efficiency is improved.
  • the fingerprint recognition method of the present invention and its respective steps will be described in detail below.
  • the fingerprint identification method in this embodiment includes a fingerprint collection step S10, a template selection step S20, and a matching step S30.
  • the fingerprint collection step S10 acquiring a fingerprint image, and extracting a fingerprint feature value from the fingerprint image.
  • the fingerprint image can be collected by using a low-speed peripheral device in the prior art, and the purpose is to extract a fingerprint feature value unique to the fingerprint image from the fingerprint image, and then compare it with the feature value of the active template extracted later.
  • the template selection step S20 selecting an active template from the template database, and extracting feature values of the active template, wherein the fingerprint template is marked with the number of matching success times, and the active template is the highest number of matching successes in the fingerprint template.
  • the fingerprint templates in the template database are marked with the number of matching successes before the matching.
  • the active template is the fingerprint template with the highest number of matching successes in the template database. Therefore, in the template selection step S20, the active template is selected from the template database.
  • Matching step S30 comparing the fingerprint feature value with the feature value of the active template; if the matching fails, returning to the execution template selection step, if the matching is successful, the number of matching success times of the active template tag is increased once. If the match has failed, the matching task is still in progress.
  • the “selected” used in the present application means that, after the active template is selected in the matching task, the active template is removed from the selection range. If the matching fails in a matching step, and the fingerprint template needs to be selected again, the remaining fingerprint templates in the selection range after the selected active template is removed from the template database are selected. That is, if the selected active template does not match successfully, the next time the matching is performed, the active template with the highest number of successful tag matching is selected from the remaining fingerprint templates in the template database. It can be seen that the active template is the template with the highest number of successful matching matches in a relative selection range, and is not a fixed template.
  • the fingerprint acquisition step S10 in the process of acquiring the fingerprint image, significant feature values are extracted from the fingerprint image. Therefore, in the process of fingerprint image acquisition, if a prominent salient feature value can be extracted, the salient feature value is extracted first. The purpose is to save time by simultaneously comparing the significant feature values with the feature values of the active templates during the fingerprint image collection process.
  • the salient feature value is compared with the feature value of the active template. For example, in the process of performing the comparison, if the matching degree does not exceed the set threshold, the feature values of the next active template selected in the template selection step are compared. If the matching degree between the significant feature value and the feature value of the active template exceeds a set threshold. For example, if the matching degree of the eighth selected active template and the significant feature value exceeds the verification threshold, after the fingerprint image acquisition is completed, all the fingerprint feature values extracted in the fingerprint image and the eighth selected active template are selected. If the matching is successful, the matching success number of the eighth active template tag is increased once. If the matching fails, the ninth active template is selected to be compared with all the fingerprint feature values extracted in the fingerprint image. The latter process can be seen as described above.
  • the matching step S30 further includes: if the matching is successful, the matching success number of the active template mark is increased once.
  • the matching success number of the active template mark is increased once.
  • the dynamic update of the matching success times of the template tags in the database can be implemented, which facilitates the rapid execution of subsequent matching tasks. Therefore, in the actual matching process, the number of matching successes is dynamically learned, and the template selection order is dynamically adjusted according to the number of matching success times of the template, so that the template with higher matching probability is preferentially compared and improved. Matching efficiency.
  • the selection algorithm of the fingerprint template with the same number of matching success times may adopt the selection algorithm in the prior art.
  • the active template is selected from the template database, including step S200: the number of matching successes marked from the template database according to the fingerprint template.
  • a plurality of fingerprint templates are sequentially selected from the most to the least to form a template set; and an active template is selected from the template set.
  • the amount of data in the template database is very large.
  • the template selection step accesses the template database, the amount of data that the terminal needs to load is relatively large, which affects the processing speed of the terminal. In this way, the active template can be selected from the template set.
  • the template database the amount of data in the template collection is much smaller. Therefore, in the template selection step, the template collection is first formed, thereby improving the execution speed of the template selection step.
  • the fingerprint identification method in this embodiment further includes a reading step S50: reading a new fingerprint template from the template database and adding a new fingerprint template to the template under the condition that the matching fails in the matching step. set.
  • the first embodiment returns to the execution template selection step S20 when the matching fails.
  • a new fingerprint template is read from the template database under the condition that the matching fails, and the new fingerprint template is added to the template set, and then the template selection step S20 is performed.
  • the purpose of this embodiment is to ensure that the number of fingerprint templates in the template set remains unchanged, and the matching thread and the reading thread are implemented in parallel, thereby improving the speed and efficiency of fingerprint recognition.
  • the new fingerprint template in this embodiment is different from the plurality of templates formed by forming the template.
  • the number of matching successes of the new fingerprint template tag is the highest among the remaining fingerprint templates of the template database. Increased the chances of a successful match.
  • the device for fingerprint recognition is a device class right corresponding to the first embodiment.
  • the device for fingerprint identification includes: a fingerprint collection module 100: for acquiring a fingerprint image, and extracting a fingerprint feature value from the fingerprint image; and a template selection module 101: for selecting an active template from the template database, and extracting the active The eigenvalue of the template, where the fingerprint template The matching template has the highest number of matching successes, and the active template is the highest number of matching successes in the fingerprint template.
  • the matching module 102 is configured to compare the fingerprint feature value with the feature value of the active template; if the matching fails, the template selection module selects again. The active template is sent out for the next match. If the match is successful, the number of matching successes of the active template tag is increased once.
  • the apparatus for fingerprint identification of the embodiment is based on the fifth embodiment, and the template selection module is further configured to: select from the template database according to the number of matching success times of the fingerprint template mark Multiple fingerprint templates are generated to form a template collection; an active template is selected from the template collection.
  • the fingerprint identification device of the embodiment further includes a reading module 104, configured to: read a new fingerprint from the template database under the condition that the matching module fails to match, on the basis of the sixth embodiment.
  • the template is added to the template collection; after the new fingerprint template is added to the template collection, the template selection module 102 selects the active template from the template collection.
  • the terminal provided in this embodiment includes the fingerprint identification device described above, thereby enabling fingerprint identification to be quickly realized, reducing user waiting time, and improving work efficiency.
  • the terminal in the embodiment of the present invention may include a communication bus 802, an input device 803, a memory 804, and a processor 801, where:
  • the communication bus 802 is configured to implement connection communication between the input device 803, the memory 804, and the processor 801.
  • the input device 803 is configured to acquire a fingerprint image.
  • the input device 803 acquires a fingerprint image, and extracts a fingerprint feature value from the fingerprint image.
  • the processor 801 selects an active template from the template database, and extracts the feature value of the active template, where the fingerprint template is marked with the number of matching success times, and the active template is the highest number of matching successes in the fingerprint template. of.
  • the processor 801 compares the fingerprint feature value with a feature value of the active template. Correct.
  • the processor 801 selects an active template from the template database, including:
  • a plurality of fingerprint templates are sequentially selected from the template database according to the number of matching success times marked by the fingerprint template to form a template set.
  • An active template is selected from the set of templates.
  • the processor 801 reads a new fingerprint template from the template database and adds the new fingerprint template to the template set under the condition that the matching fails.
  • the number of matching success times of the new fingerprint template tag is the highest among the remaining fingerprint templates of the template database.
  • the method further includes:
  • the input device 803 extracts a significant feature value from the fingerprint image during the process of acquiring the fingerprint image.
  • the processor 801 compares the fingerprint feature value with the feature value of the active template, including:
  • the feature values of the next active template selected in the template selection step are compared.
  • the matching degree exceeds the set threshold, compare all the fingerprint feature values extracted after acquiring the fingerprint image in the fingerprint collecting step with the feature values of the active template, and if the matching fails, return to execute the template selecting step, if If the match is successful, the number of matching successes of the active template tag is increased once.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Pathology (AREA)
  • Theoretical Computer Science (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Dentistry (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Collating Specific Patterns (AREA)

Abstract

一种指纹识别的方法、装置和终端。该指纹识别的方法包括：指纹采集步骤（S10）：获取指纹图像，从所述指纹图像中提取指纹特征值；模板选取步骤(S20)：从模板数据库中选出活跃模板，并提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；匹配步骤(S30)：将所述指纹特征值与所述活跃模板的特征值进行比对；若匹配失败，返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次，提高了指纹识别的速度。

Description

一种指纹识别方法、装置和终端
本申请要求于2015年06月30日提交中国专利局，申请号为201510390466.1、发明名称为“一种指纹识别方法、装置和终端”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及指纹识别技术领域，尤其涉及一种指纹识别方法、装置和终端。
背景技术
目前，指纹识别的流程为：首先使用指纹模组进行指纹模板采集；然后对指纹模板进行初步处理后存储；接着采集指纹并与指纹模板进行特征比对；最后输出比对结果。
可以看出，指纹识别的关键技术为采集指纹并与模板进行比对的过程，对于高安全、高精密的领域中只有一个模板的情况比较少见。通常存在的场景是：一个指纹与大量模板进行匹配，比如考勤、门禁、公安破案等。由于模板库中存在大量的模板，比对的过程耗时较长，很难被人接受。因此，需要一种指纹识别方法、装置和终端，以解决现有技术中存在的上述技术问题。
发明内容
本发明要解决的技术问题是，提供一种指纹识别方法、装置和终端，提高了指纹识别的速度。
本发明采用的技术方案是：
一种指纹识别的方法，其包括：指纹采集步骤：获取指纹图像，从所述指纹图像中提取指纹特征值；模板选取步骤：从模板数据库中选出活跃模板，并提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；匹配步骤：将所述指纹特征值与所述活跃模板的特征值进行比对；若匹配失败， 返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
优选地，在模板选取步骤中，所述从模板数据库中选出活跃模板，包括：从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；从所述模板集合中选出活跃模板。
优选地，所述方法还包括：读取步骤：在所述匹配步骤中匹配失败的条件下，从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合；将所述新指纹模板加入所述模板集合后，再从所述模板集合中选出活跃模板。
优选地，所述新指纹模板标记的匹配成功次数是所述模板数据库中剩余指纹模板中最高的。
优选地，所述指纹采集步骤，还包括：获取指纹图像的过程中从所述指纹图像中提取显著特征值；所述匹配步骤，包括：将所述显著特征值与所述活跃模板的特征值比对；若匹配度未超过设定阈值，与模板选取步骤中选出的下一活跃模板的特征值比对；若匹配度超过设定阈值，将所述指纹采集步骤中获取指纹图像后提取的所有指纹特征值与所述活跃模板的特征值进行比对，若匹配失败，返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
本发明还提供了一种指纹识别的装置，其包括：指纹采集模块：用于获取指纹图像，以从所述指纹图像提取指纹特征值；模板选取模块：用于从模板数据库中选出活跃模板，以提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；匹配模块：用于将所述指纹特征值与所述活跃模板的特征值进行比对；若匹配失败，模板选取模块再次选出活跃模板，以进行下一次匹配，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
优选地，所述模板选取模块，还用于：从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；从所述模板集合中选出活跃模板。
优选地，所述装置还包括读取模块，用于：在匹配模块匹配失败的条 件下，从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合；将所述新指纹模板加入所述模板集合后，所述模板选取模块从所述模板集合中选出活跃模板。
此外，本发明还提供一种终端，包括通信总线、输入装置、存储器以及处理器，其中：
所述通信总线，用于实现所述输入装置、存储器以及处理器之间的连接通信；
所述输入装置，用于获取指纹图像；
所述存储器中存储一组程序代码，且所述终端调用所述存储器中存储的程序代码，用于执行以下操作：
所述输入装置获取指纹图像，从所述指纹图像中提取指纹特征值；
所述处理器从模板数据库中选出活跃模板，并提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；
所述处理器将所述指纹特征值与所述活跃模板的特征值进行比对；
若匹配失败，返回执行所述模板选取步骤，若匹配成功，则所述处理器为所述活跃模板标记的匹配成功次数增加一次。
优选地，所述处理器从模板数据库中选出活跃模板，包括：
从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；
从所述模板集合中选出活跃模板。
优选地，还包括：
在所述匹配失败的条件下，所述处理器从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合；
所述处理器将所述新指纹模板加入所述模板集合后，再从所述模板集合中选出活跃模板。
优选地，所述新指纹模板标记的匹配成功次数是所述模板数据库的剩余指纹模板中最高的。
优选地，还包括：
所述输入装置获取指纹图像的过程中从所述指纹图像中提取显著特征值；
所述处理器将所述指纹特征值与所述活跃模板的特征值进行比对，包括：
将所述显著特征值与所述活跃模板的特征值比对；
若匹配度未超过设定阈值，与模板选取步骤中选出的下一活跃模板的特征值比对；
若匹配度超过设定阈值，将所述指纹采集步骤中获取指纹图像后提取的所有指纹特征值与所述活跃模板的特征值进行比对，若匹配失败，返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
采用上述技术方案，本发明至少具有下列效果：
本发明所述指纹识别方法提高了指纹识别的速度以及效率，减少了用户在进行指纹识别时等待的时间。
附图说明
图1为本发明第一实施例的指纹识别方法的流程图；
图2为本发明第二实施例的指纹识别方法的流程图；
图3为本发明第三实施例的指纹识别方法的流程图；
图4为本发明第四实施例的指纹识别方法的流程图；
图5为本发明的第五实施例的指纹识别装置的方框图；
图6为本发明的第六实施例的指纹识别装置的方框图；
图7为本发明的第七实施例的指纹识别装置的方框图；
图8为本发明的第八实施例的终端的方框图。
具体实施方式
为更进一步阐述本发明为达成预定目的所采取的技术手段及功效，以下结合附图及较佳实施例，对本发明进行详细说明如后。
相对于现有技术中的指纹识别方法，本发明提供的指纹识别的方法可 以较快地完成指纹的识别，提高了工作效率。下面将详细地描述本发明的指纹识别方法及其各个步骤。
第一实施例
如图1所示，本实施例中的指纹识别方法包括指纹采集步骤S10、模板选取步骤S20以及匹配步骤S30。
其中，指纹采集步骤S10：获取指纹图像，从指纹图像提取指纹特征值。可以采用现有技术中的低速外设进行指纹图像的采集，目的是从指纹图像中提取表征该指纹图像独一无二的指纹特征值，然后与后面提取的活跃模板的特征值进行比对。
模板选取步骤S20：从模板数据库中选出活跃模板，并提取活跃模板的特征值，其中指纹模板均标记有匹配成功次数，活跃模板为指纹模板中标记的匹配成功次数最高的。区别于现有技术的是，模板数据库中的指纹模板均标记有在此之前进行匹配的匹配成功次数。而活跃模板是该模板数据库中标记的匹配成功次数最高的指纹模板。因此，在模板选取步骤S20中，从模板数据库中选出该活跃模板。
需要说明的是，步骤S10和步骤S20没有明显的先后顺序，在此仅是为了表述的方便。
匹配步骤S30：将指纹特征值与活跃模板的特征值进行比对；若匹配失败，返回执行模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。如果匹配一直失败，则该匹配任务就一直处于进行中。
在本申请中使用的“选出”是指，在该次匹配任务中，将活跃模板选出之后，就将该活跃模板从选择范围中剔除。如果在一次匹配步骤中匹配失败，需要再一次选取指纹模板，那么从模板数据库中剔除此次选出的活跃模板后的选择范围内的剩余指纹模板中进行选取。也就是，若选出的活跃模板没有匹配成功，那么下一次进行匹配时，就从模板数据库的剩余指纹模板中选出标记匹配成功次数最高的活跃模板。可以看出，活跃模板是在一个相对的选取范围中标记匹配成功次数最高的模板，并不是一个固定模板。
作为优选的实施例，指纹采集步骤S10中，获取指纹图像的过程中，从指纹图像中提取显著特征值。因此，在指纹图像采集的过程中，如果可以提取到较为突出的显著特征值，那么先提取该显著特征值。目的是，在指纹图像的采集过程中，同时进行显著特征值与活跃模板的特征值的比对，以此节省时间。
从而，在匹配步骤中：将显著特征值与活跃模板的特征值比对。举例说明，在进行比对的过程中，若匹配度未超过设定阈值，与模板选取步骤中选出的下一活跃模板的特征值比对。若显著特征值与活跃模板的特征值的匹配度超过设定阈值。例如可以是第八次选出的活跃模板与显著特征值的匹配度超过审定阈值，那么在指纹图像获取完成之后，将指纹图像中提取的所有指纹特征值与该第八次选出的活跃模板进行比对，如果匹配成功，则为第八个活跃模板标记的匹配成功次数增加一次，如果匹配失败，则选出第九个活跃模板来与该指纹图像中提取的所有指纹特征值进行比对，后面的过程可以参见上文所述。
第二实施例
本实施例在第一实施例的基础上，匹配步骤S30还包括：若匹配成功，将活跃模板标记的匹配成功次数增加一次。由此，只要匹配成功一次，就可以将活跃模板标记的匹配成功次数增加一次。可以实现数据库中模板标记的匹配成功次数的动态更新，便于后续匹配任务的快速执行。由此，在实际的匹配过程中，对匹配成功次数进行动态的学习，根据模板的匹配成功次数，对模板的选出顺序进行动态的调整，使得匹配概率较高的模板优先完成比对，提高匹配的效率。另外，需要说明的是，对于匹配成功次数相同的指纹模板的选出顺序可以采用现有技术中的选出算法。
如图2中以模板数据库中包括5个模板的情况进行说明。处于最上面的是模板数据库中从模板1至模板5的选出顺序，其中，模板1处于最先选出的位置，此时，所有模板标记的匹配成功次数均为0。随着执行一次匹配步骤后，模板5匹配成功，那么相应的模板5标记的匹配成功次数为1，并且模板5排在选出位置的第一位。当完成n次匹配步骤后，相应的模板2标记的匹配成功次数为a，相应的模板5标记的匹配成功次数为b，相 应的模板3标记的匹配成功次数为c，相应的模板1标记的匹配成功次数为D，相应的模板4标记的匹配成功次数为E。并且a>b>c>D>E，由此获得此时的选出顺序。那么在执行第n+1次匹配步骤时，首先选出模板2进行比对，如果匹配失败，再选出模板5，直至结束。
第三实施例
如图3所示，在第一实施例的基础上，本实施例在在模板选取步骤中，从模板数据库中选出活跃模板，包括步骤S200：从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；从模板集合中选出活跃模板。模板数据库的数据量非常大，每次模板选取步骤访问该模板数据库时，终端需要加载的数据量比较大，影响终端的处理速度，而采用这种方式可以从模板集合中选出活跃模板，相对于模板数据库，模板集合的数据量要小很多，因此，在模板选取步骤中，首先形成模板集合，以此提高了模板选取步骤的执行速度。
第四实施例
如图4所示，本实施例中的指纹识别方法还包括读取步骤S50：在匹配步骤中匹配失败的条件下，从模板数据库中读取一个新指纹模板，并且将新指纹模板加入到模板集合。第一实施例是在匹配失败时，返回执行模板选取步骤S20。相比于第一实施例，本实施例是在匹配失败的条件下，从模板数据库中读取一个新指纹模板，并且将该新指纹模板加入到模板集合中，然后，再执行模板选取步骤S20。本实施例的目的是确保模板集合中的指纹模板数量保持不变，实现匹配线程与读取线程并行进行，提高指纹识别的速度和效率。可以看出，本实施例中的新指纹模板是区别于上述形成模板集成的多个模板。优选地，新指纹模板标记的匹配成功次数在模板数据库的剩余指纹模板中最高的。提高了匹配成功的机率。
第五实施例
如图5所示，本实施例提供的指纹识别的装置是对应于第一实施例的装置类权项。如图所示，该指纹识别的装置包括：指纹采集模块100：用于获取指纹图像，从指纹图像提取指纹特征值；模板选取模块101：用于从模板数据库中选出活跃模板，并提取活跃模板的特征值，其中指纹模板 均标记有匹配成功次数，活跃模板为指纹模板中标记的匹配成功次数最高的；匹配模块102：用于将指纹特征值与活跃模板的特征值进行比对；若匹配失败，模板选取模块再次选出活跃模板，以进行下一次匹配，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
第六实施例
如图6所述，本实施例的指纹识别的装置在第五实施例的基础上，模板选取模块，还用于：从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；从模板集合中选出活跃模板。
第七实施例
如图7所示，本实施例的指纹识别装置在第六实施例的基础上，还包括读取模块104，用于：在匹配模块匹配失败的条件下，从模板数据库中读取一个新指纹模板，并且将新指纹模板加入到模板集合；将新指纹模板加入模板集合后，模板选取模块102从模板集合中选出活跃模板。
第八实施例
本实施例提供的终端包括上文所述的指纹识别装置，由此可以快速地实现指纹的识别，减少用户等待时间，提高工作效率。
如图8所示，本发明实施例中的终端可以包括通信总线802、输入装置803、存储器804以及处理器801，其中：
所述通信总线802，用于实现所述输入装置803、存储器804以及处理器801之间的连接通信。
所述输入装置803，用于获取指纹图像。
所述存储器804中存储一组程序代码，且所述终端调用所述存储器804中存储的程序代码，用于执行以下操作：
所述输入装置803获取指纹图像，从所述指纹图像中提取指纹特征值。
所述处理器801从模板数据库中选出活跃模板，并提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的。
所述处理器801将所述指纹特征值与所述活跃模板的特征值进行比 对。
若匹配失败，返回执行所述模板选取步骤，若匹配成功，则所述处理器801为所述活跃模板标记的匹配成功次数增加一次。
在可选实施例中，处理器801从模板数据库中选出活跃模板，包括：
从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合。
从所述模板集合中选出活跃模板。
在可选实施例中，还包括：
在所述匹配失败的条件下，所述处理器801从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合。
所述处理器801将所述新指纹模板加入所述模板集合后，再从所述模板集合中选出活跃模板。
在可选实施例中，所述新指纹模板标记的匹配成功次数是所述模板数据库的剩余指纹模板中最高的。
在可选实施例中，还包括：
所述输入装置803获取指纹图像的过程中从所述指纹图像中提取显著特征值。
所述处理器801将所述指纹特征值与所述活跃模板的特征值进行比对，包括：
将所述显著特征值与所述活跃模板的特征值比对。
若匹配度未超过设定阈值，与模板选取步骤中选出的下一活跃模板的特征值比对。
若匹配度超过设定阈值，将所述指纹采集步骤中获取指纹图像后提取的所有指纹特征值与所述活跃模板的特征值进行比对，若匹配失败，返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
通过具体实施方式的说明，应当可对本发明为达成预定目的所采取的技术手段及功效得以更加深入且具体的了解，然而所附图示仅是提供参考与说明之用，并非用来对本发明加以限制。

Claims (13)

  1. 一种指纹识别的方法，其特征在于，包括：
    指纹采集步骤：获取指纹图像，从所述指纹图像中提取指纹特征值；
    模板选取步骤：从模板数据库中选出活跃模板，并提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；
    匹配步骤：将所述指纹特征值与所述活跃模板的特征值进行比对；
    若匹配失败，返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
  2. 根据权利要求1所述的方法，其特征在于，在模板选取步骤中，所述从模板数据库中选出活跃模板，包括：
    从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；
    从所述模板集合中选出活跃模板。
  3. 根据权利要求2所述的方法，其特征在于，所述方法还包括：
    读取步骤：在所述匹配步骤中匹配失败的条件下，从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合；
    将所述新指纹模板加入所述模板集合后，再从所述模板集合中选出活跃模板。
  4. 根据权利要求3所述的方法，其特征在于，所述新指纹模板标记的匹配成功次数是所述模板数据库的剩余指纹模板中最高的。
  5. 根据权利要求1所述的方法，其特征在于，所述指纹采集步骤，还包括：获取指纹图像的过程中从所述指纹图像中提取显著特征值；
    所述匹配步骤，包括：
    将所述显著特征值与所述活跃模板的特征值比对；
    若匹配度未超过设定阈值，与模板选取步骤中选出的下一活跃模板的特征值比对；
    若匹配度超过设定阈值，将所述指纹采集步骤中获取指纹图像后提取 的所有指纹特征值与所述活跃模板的特征值进行比对，若匹配失败，返回执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
  6. 一种指纹识别的装置，其特征在于，包括：
    指纹采集模块：用于获取指纹图像，以从所述指纹图像提取指纹特征值；
    模板选取模块：用于从模板数据库中选出活跃模板，以提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；
    匹配模块：用于将所述指纹特征值与所述活跃模板的特征值进行比对；
    若匹配失败，模板选取模块再次选出活跃模板，以进行下一次匹配，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
  7. 根据权利要求6所述的装置，其特征在于，所述模板选取模块，还用于：
    从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；
    从所述模板集合中选出活跃模板。
  8. 根据权利要求7所述的装置，其特征在于，所述装置还包括读取模块，用于：在匹配模块匹配失败的条件下，从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合；
    将所述新指纹模板加入所述模板集合后，所述模板选取模块从所述模板集合中选出活跃模板。
  9. 一种终端，其特征在于，包括通信总线、输入装置、存储器以及处理器，其中：
    所述通信总线，用于实现所述输入装置、存储器以及处理器之间的连接通信；
    所述输入装置，用于获取指纹图像；
    所述存储器中存储一组程序代码，且所述终端调用所述存储器中存储的程序代码，用于执行以下操作：
    所述输入装置获取指纹图像，从所述指纹图像中提取指纹特征值；
    所述处理器从模板数据库中选出活跃模板，并提取所述活跃模板的特征值，其中指纹模板均标记有匹配成功次数，所述活跃模板为所述指纹模板中标记的匹配成功次数最高的；
    所述处理器将所述指纹特征值与所述活跃模板的特征值进行比对；
    若匹配失败，返回执行所述模板选取步骤，若匹配成功，则所述处理器为所述活跃模板标记的匹配成功次数增加一次。
  10. 根据权利要求9所述的终端，其特征在于，所述处理器从模板数据库中选出活跃模板，包括：
    从模板数据库中根据指纹模板标记的匹配成功次数由多到少顺次选出多个指纹模板，以形成模板集合；
    从所述模板集合中选出活跃模板。
  11. 根据权利要求10所述的终端，其特征在于，还包括：
    在所述匹配失败的条件下，所述处理器从所述模板数据库中读取一个新指纹模板，并且将所述新指纹模板加入到所述模板集合；
    所述处理器将所述新指纹模板加入所述模板集合后，再从所述模板集合中选出活跃模板。
  12. 根据权利要求11所述的终端，其特征在于，所述新指纹模板标记的匹配成功次数是所述模板数据库的剩余指纹模板中最高的。
  13. 根据权利要求9所述的终端，其特征在于，还包括：
    所述输入装置获取指纹图像的过程中从所述指纹图像中提取显著特征值；
    所述处理器将所述指纹特征值与所述活跃模板的特征值进行比对，包括：
    将所述显著特征值与所述活跃模板的特征值比对；
    若匹配度未超过设定阈值，与模板选取步骤中选出的下一活跃模板的特征值比对；
    若匹配度超过设定阈值，将所述指纹采集步骤中获取指纹图像后提取的所有指纹特征值与所述活跃模板的特征值进行比对，若匹配失败，返回 执行所述模板选取步骤，若匹配成功，则为所述活跃模板标记的匹配成功次数增加一次。
PCT/CN2015/085843 2015-06-30 2015-07-31 一种指纹识别方法、装置和终端 Ceased WO2017000352A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510390466.1 2015-06-30
CN201510390466.1A CN106326819B (zh) 2015-06-30 2015-06-30 一种指纹识别方法、装置和终端

Publications (1)

Publication Number Publication Date
WO2017000352A1 true WO2017000352A1 (zh) 2017-01-05

Family

ID=57607573

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/085843 Ceased WO2017000352A1 (zh) 2015-06-30 2015-07-31 一种指纹识别方法、装置和终端

Country Status (2)

Country Link
CN (1) CN106326819B (zh)
WO (1) WO2017000352A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106845198A (zh) * 2017-01-23 2017-06-13 宇龙计算机通信科技(深圳)有限公司 一种指纹验证的方法和装置
CN106909900A (zh) * 2017-02-27 2017-06-30 努比亚技术有限公司 指纹识别方法及装置
CN110569748A (zh) * 2019-08-21 2019-12-13 杭州指安科技股份有限公司 一种指纹识别系统中快速反馈比对结果的电子装置及方法
CN110853676B (zh) * 2019-11-18 2023-06-09 广州国音智能科技有限公司 一种音频比对方法、装置及设备
CN112528946A (zh) * 2020-12-24 2021-03-19 北京深思数盾科技股份有限公司 一种指纹识别方法、装置、存储介质及电子设备

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1439997A (zh) * 2002-02-22 2003-09-03 杭州中正生物认证技术有限公司 指纹识别方法与系统
KR20050031346A (ko) * 2003-09-29 2005-04-06 (주)위트정보통신 지문 분류 및 인식을 위한 입력장치 및 방법
CN1818927A (zh) * 2006-03-23 2006-08-16 北京中控科技发展有限公司 指纹识别方法与系统
US7155039B1 (en) * 2002-12-18 2006-12-26 Motorola, Inc. Automatic fingerprint identification system and method
CN103336800A (zh) * 2013-06-19 2013-10-02 武汉世纪金桥安全技术有限公司 基于行为分析的指纹存储比对的方法
CN103345616A (zh) * 2013-06-19 2013-10-09 武汉世纪金桥安全技术有限公司 基于行为分析的指纹存储比对的系统
CN104463141A (zh) * 2014-12-24 2015-03-25 北京大唐智能卡技术有限公司 一种基于智能卡的指纹模板录入方法及装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101751550B (zh) * 2008-12-19 2012-02-01 杭州中正生物认证技术有限公司 快速指纹搜索方法及快速指纹搜索系统
TW201419011A (zh) * 2012-11-06 2014-05-16 Chunghwa Wideband Best Network Co Ltd 個人化選單呈現方法及系統
CN104036268A (zh) * 2014-07-03 2014-09-10 南昌欧菲生物识别技术有限公司 指纹注册方法、快速指纹认证方法及终端设备

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1439997A (zh) * 2002-02-22 2003-09-03 杭州中正生物认证技术有限公司 指纹识别方法与系统
US7155039B1 (en) * 2002-12-18 2006-12-26 Motorola, Inc. Automatic fingerprint identification system and method
KR20050031346A (ko) * 2003-09-29 2005-04-06 (주)위트정보통신 지문 분류 및 인식을 위한 입력장치 및 방법
CN1818927A (zh) * 2006-03-23 2006-08-16 北京中控科技发展有限公司 指纹识别方法与系统
CN103336800A (zh) * 2013-06-19 2013-10-02 武汉世纪金桥安全技术有限公司 基于行为分析的指纹存储比对的方法
CN103345616A (zh) * 2013-06-19 2013-10-09 武汉世纪金桥安全技术有限公司 基于行为分析的指纹存储比对的系统
CN104463141A (zh) * 2014-12-24 2015-03-25 北京大唐智能卡技术有限公司 一种基于智能卡的指纹模板录入方法及装置

Also Published As

Publication number Publication date
CN106326819B (zh) 2019-11-12
CN106326819A (zh) 2017-01-11

Similar Documents

Publication Publication Date Title
CN104463141B (zh) 一种基于智能卡的指纹模板录入方法及装置
US10108792B2 (en) Biometric-based authentication method, apparatus and system
WO2017008349A1 (zh) 指纹识别的方法及装置
WO2017000352A1 (zh) 一种指纹识别方法、装置和终端
US20150199568A1 (en) Automated document recognition, identification, and data extraction
US20210216617A1 (en) Biometric authentication device, biometric authentication method, and computer-readable recording medium recording biometric authentication program
CN110069994A (zh) 基于人脸多区域的人脸属性识别系统、方法
JP2011521355A (ja) 勾配ヒストグラムを用いた指紋表現
CN111401374A (zh) 基于多任务的模型训练方法、字符识别方法及装置
WO2018090641A1 (zh) 识别保险单号码的方法、装置、设备及计算机可读存储介质
CN111241930A (zh) 一种用于人脸识别的方法及系统
CN105022984A (zh) 指纹采集方法、指纹比对方法、指纹识别装置和系统
WO2016201731A1 (zh) 一种指纹识别方法、装置及电子设备
US8842917B2 (en) Local feature extraction apparatus, control method therefor, and computer-readable medium
CN110390353B (zh) 一种基于图像处理的生物识别方法和系统
US20080085039A1 (en) Fingerprint matching method and apparatus using fingerprint minutiae and fingerprint binary images
JP2013105373A (ja) データ取得装置、方法及びプログラム
CN111062301A (zh) 身份认证方法、装置、电子设备及计算机可读存储介质
CN109190676B (zh) 一种用于图像识别的模型训练方法、装置、设备及存储介质
CN111222367A (zh) 一种指纹识别方法及装置、存储介质、终端
CN115116073A (zh) 洗涤标签识别方法、装置、存储介质及电子装置
CN112016466A (zh) 人脸识别方法、系统、电子设备和计算机存储介质
JPWO2022014062A5 (ja) 価値譲渡装置、価値譲渡システム、価値譲渡方法及び価値譲渡プログラム
CN111310548A (zh) 一种在线手写笔迹中笔画类型的识别方法
WO2024159914A1 (zh) 弱监督时序边界定位方法、装置、电子设备及存储介质

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: 15896882

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: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 17/05/2018)

122 Ep: pct application non-entry in european phase

Ref document number: 15896882

Country of ref document: EP

Kind code of ref document: A1