TWI822646B - Identity verification device and method based on dynamic lip image - Google Patents
Identity verification device and method based on dynamic lip image Download PDFInfo
- Publication number
- TWI822646B TWI822646B TW112124001A TW112124001A TWI822646B TW I822646 B TWI822646 B TW I822646B TW 112124001 A TW112124001 A TW 112124001A TW 112124001 A TW112124001 A TW 112124001A TW I822646 B TWI822646 B TW I822646B
- Authority
- TW
- Taiwan
- Prior art keywords
- dynamic image
- lip
- operator
- similarity
- voice message
- Prior art date
Links
- 238000012795 verification Methods 0.000 title claims abstract description 93
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000008859 change Effects 0.000 claims description 28
- 230000008569 process Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 18
- 230000006870 function Effects 0.000 description 4
- 230000009977 dual effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Collating Specific Patterns (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
Description
本發明係關於一種身分驗證裝置及方法,特別是一種應用唇形動態影像及語音分析的身分驗證裝置及方法。The present invention relates to an identity verification device and method, particularly an identity verification device and method using lip dynamic images and voice analysis.
隨著生物辨識技術的蓬勃發展,現今眾多產業(例如銀行交易)時常應用指紋辨識或臉部辨識作為客戶身份辨識的依據。針對前者來說,僅有新款的裝置支援指紋辨識功能,對於舊型的裝置來說並無法提供指紋辨識功能。針對後者來說,不同個體之間區別不大,且同樣的個體可能因外在樣貌改變而導致無法準確地辨識。With the vigorous development of biometric recognition technology, many industries (such as banking transactions) now often use fingerprint recognition or facial recognition as the basis for customer identification. For the former, only newer devices support the fingerprint recognition function, and older devices cannot provide the fingerprint recognition function. For the latter, there is not much difference between different individuals, and the same individual may not be accurately identified due to changes in external appearance.
對於注重客戶身份確認的特定行業來說,若是因客戶身份辨識有瑕疵而導致金融交易出現錯誤,例如客戶身分遭盜用,不但可能會使客戶蒙受金錢損失,同時也會影響企業的商譽。因此,在這類型的業界需要一種用戶的身分辨識裝置,其辨識方式可兼顧準確性與便利性,以利用戶可以安全且有效率地執行交易或業務。For certain industries that focus on customer identity verification, if there are errors in financial transactions due to flaws in customer identity recognition, such as customer identity theft, it may not only cause customers to suffer monetary losses, but also affect the goodwill of the company. Therefore, in this type of industry, there is a need for a user's identity recognition device whose recognition method can take into account both accuracy and convenience, so that users can safely and efficiently perform transactions or business.
有鑑於此,本發明提出一種基於動態影像的身分驗證裝置及方法,透過語音辨識搭配唇形動態影像分析,以雙重認證的方式確認用戶身份,如此可大大提升用戶身份辨識的準確性及用戶使用上的便利性。In view of this, the present invention proposes an identity verification device and method based on dynamic images. Through voice recognition and lip dynamic image analysis, the user's identity is confirmed in a double authentication method. This can greatly improve the accuracy of user identity recognition and user usage. convenience.
依據本發明之一實施例揭露一種基於動態影像的身分驗證裝置,包含錄像器、資料庫及運算器。錄像器用以依據控制指令擷取包含第一語音訊息的第一唇部動態影像,其中第一語音訊息關聯於待驗證個資。資料庫用於儲存包含第二語音訊息的第二唇部動態影像及預設個資。運算器連接錄像器及資料庫,運算器用以依據一請求指令而產生控制指令,且運算器判斷待驗證個資是否符合預設個資。當待驗證個資符合預設個資時,運算器進一步判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值,並在相似度大於驗證閾值時產生驗證成功訊息。According to an embodiment of the present invention, an identity verification device based on dynamic images is disclosed, which includes a video recorder, a database and a computing unit. The video recorder is used to capture the first lip dynamic image including the first voice message according to the control instruction, wherein the first voice message is associated with the personal information to be verified. The database is used to store the second lip dynamic image including the second voice message and the default personal information. The operator is connected to the video recorder and the database. The operator is used to generate control instructions according to a request instruction, and the operator determines whether the information to be verified matches the preset information. When the personal information to be verified matches the preset personal information, the operator further determines whether the similarity between the first lip dynamic image and the second lip dynamic image is greater than the verification threshold, and generates a verification success message when the similarity is greater than the verification threshold.
依據本發明之一實施例揭露一種基於動態影像的身分驗證方法,包含以一運算器依據關聯於一預設個資的一請求指令而產生一控制指令;以錄像器依據該控制指令擷取包含一第一語音訊息的一第一唇部動態影像,其中該第一語音訊息關聯於一待驗證個資;以運算器判斷該待驗證個資是否符合該預設個資;當該待驗證個資符合該預設個資時,以該運算器判斷該第一唇部動態影像與一資料庫內的一第二唇部動態影像的一相似度是否大於一驗證閾值,其中該第二唇部動態影像包含一第二語音訊息;以及當該相似度大於該驗證閾值時,以該運算器產生一驗證成功訊息。According to an embodiment of the present invention, an identity verification method based on dynamic images is disclosed, which includes using a computer to generate a control command based on a request command associated with a preset data; using a video recorder to capture the content containing A first lip dynamic image of a first voice message, wherein the first voice message is associated with a profile to be verified; using a computer to determine whether the profile to be verified matches the default profile; when the profile to be verified is When the preset personal information is met, the operator is used to determine whether a similarity between the first lip dynamic image and a second lip dynamic image in a database is greater than a verification threshold, wherein the second lip dynamic image The dynamic image includes a second voice message; and when the similarity is greater than the verification threshold, the operator is used to generate a verification success message.
綜上所述,在本發明的基於動態影像的身分驗證裝置及方法中,先是利用語音訊息分析其語意內容以比對到預設個資作為第一道驗證,待通過此第一道驗證後,接著使用擷取到的唇部動態影像與預設的唇部動態影像進行相似度比對以作為第二道驗證,進而在通過兩道驗證時產生驗證成功訊息。如此一來,以雙重的語音及動態影像認證的方式重複確認用戶身份,可顯著提升用戶身份辨識的準確性且亦增加用戶的驗證過程的便利性。To sum up, in the dynamic image-based identity verification device and method of the present invention, the semantic content of the voice message is first analyzed to match the preset personal information as the first verification. After passing this first verification , and then use the captured lip dynamic image to compare the similarity with the preset lip dynamic image as the second verification, and then generate a verification success message when the two verifications are passed. In this way, repeatedly confirming the user's identity through dual voice and dynamic image authentication can significantly improve the accuracy of user identity recognition and also increase the convenience of the user's verification process.
以上之關於本揭露內容之說明及以下之實施方式之說明係用以示範與解釋本發明之精神與原理,並且提供本發明之專利申請範圍更進一步之解釋。The above description of the present disclosure and the following description of the embodiments are used to demonstrate and explain the spirit and principles of the present invention, and to provide further explanation of the patent application scope of the present invention.
以下在實施方式中詳細敘述本發明之詳細特徵以及優點,其內容足以使任何熟習相關技藝者了解本發明之技術內容並據以實施,且根據本說明書所揭露之內容、申請專利範圍及圖式,任何熟習相關技藝者可輕易地理解本發明相關之目的及優點。以下之實施例係進一步詳細說明本發明之觀點,但非以任何觀點限制本發明之範疇。The detailed features and advantages of the present invention are described in detail below in the implementation mode. The content is sufficient to enable anyone skilled in the relevant art to understand the technical content of the present invention and implement it according to the content disclosed in this specification, the patent scope and the drawings. , anyone familiar with the relevant art can easily understand the relevant objectives and advantages of the present invention. The following examples further illustrate the aspects of the present invention in detail, but do not limit the scope of the present invention in any way.
請參照圖1,圖1係依據本發明之一實施例所繪示的基於動態影像的身分驗證裝置的功能方塊圖。如圖1所示,身分驗證裝置1包含錄像器10、資料庫11及運算器12,其中運算器12分別連接錄像器10與資料庫11。錄像器10用以依據控制指令擷取包含第一語音訊息的第一唇部動態影像。所述的控制指令係來自運算器12根據一請求指令而產生的。在實作上,請求指令可以係來自用戶在裝置上登入網路銀行帳戶後所提出的交易請求,也可以是來自用戶在自動櫃員機台插卡進入銀行帳戶後所提出的交易請求,其中請求指令可例如是轉帳、匯款或提存等金融交易活動之請求指令,但本發明不以此為限。資料庫11儲存包含第二語音訊息的第二唇部動態影像以及預設個資,其中所述具有第二語音訊息的第二唇部動態影像以及預設個資係預先取得並儲存在資料庫11作為比對的依據。其中,雖然圖中未繪示,資料庫11係位於身分驗證裝置1的記憶體或儲存媒介當中。在實作上,資料庫11中內存的具有第二語音訊息的第二唇部動態影像與預設個資係依據登入之銀行帳戶的擁有者的資料(包含唇部動態影像資料與文字資料)所預先設定的。Please refer to FIG. 1 , which is a functional block diagram of an identity verification device based on dynamic images according to an embodiment of the present invention. As shown in FIG. 1 , the
錄像器10所擷取的第一語音訊息係關聯於待驗證個資,而運算器12可判斷此待驗證個資是否符合資料庫11內的預設個資。具體來說,錄像器10會將所擷取到的具有第一語音訊息之第一唇部動態影像傳送到運算器12。透過外部的語音轉文字伺服器 (例如Google Cloud語音轉文字伺服器)或是內建的功能晶片,運算器12可對第一語音訊息解析出對應的文字訊息作為待驗證個資(例如用戶的身分證字號A123456789)。The first voice message captured by the
接著,運算器12判斷此待驗證個資是否符合預設個資,若兩者符合,運算器12進一步判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值。當相似度大於驗證閾值時,運算器12產生驗證成功訊息。也就是說,在此實施例中,運算器12係先判斷待驗證個資內的文字訊息是否與預設個資內的文字訊息相同(例如身分證字號A123456789)。若是,運算器12才會進一步分析用戶當前的唇形變化與預設的唇形變化之間的相似程度,最終據以判斷是否驗證成功。在實務上,運算器12可例如是處理器、微處理器、控制器或是其他具有運算功能的晶片。Then, the
於一實施例中,第一語音訊息對應具有多個第一單位時間的第一時序,而第二語音訊息對應具有多個第二單位時間的第二時序。詳細來說,請進一步參照圖2,圖2係依據本發明之一實施例所繪示的第一語音訊息及第二語音訊息的示意圖。如圖2所示,於本發明的一實施例中,運算器12擷取第一語音訊息DA1的第一音節資料d1~d5所對應的時序以做為第一時序P1,且運算器12擷取第二語音訊息DA2的第二音節資料d1’~d5’所對應的時序作為第二時序P2。詳言之,第一語音訊息DA1具有第一音節資料d1~d5分別對應到第一時序P1的多個第一單位時間t1~t5,而第二語音訊息DA2具有第二音節資料d1’~d5’分別對應到第二時序P2的多個第二單位時間t1’~t5’。在此實施例中,第一音節資料d1~d5與第二音節資料d1’~d5’個別對應指示相同的單字或單詞。舉例來說,第一音節資料d1與第二音節資料d1’ 指示相同的單字或單詞,第一音節資料d2與第二音節資料d2’ 指示相同的單字或單詞,第一音節資料d3與第二音節資料d3’ 指示相同的單字或單詞,後續依此類推。In one embodiment, the first voice message corresponds to a first time sequence having a plurality of first unit times, and the second voice message corresponds to a second time sequence having a plurality of second unit times. For details, please further refer to FIG. 2 , which is a schematic diagram of a first voice message and a second voice message according to an embodiment of the present invention. As shown in Figure 2, in an embodiment of the present invention, the
於一實施例中,運算器12在該些第一單位時間的第N個偵測第一唇部動態影像的第一特徵點,以計算第一特徵點在第N個第一單位時間內的第一距離變化量。進一步地,運算器12在該些第二單位時間的第N個偵測第二唇部動態影像的第二特徵點,以計算第二特徵點在第N個第二單位時間內的第二距離變化量,其中N為正整數。接著,運算器12便可根據計算所得之第一距離變化量及第二距離變化量,判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值。In one embodiment, the
以下將以第一實際範例說明上述運算器12係如何判斷第一唇部動態影像與第二唇部動態影像之間的相似度是否大於驗證閾值的具體實施方式。請進一步參照圖3A與圖3B,圖3A係依據本發明之一實施例所繪示的用戶個體的第一唇部動態影像示意圖,而圖3B係依據本發明之一實施例所繪示的預存之第二唇部動態影像示意圖。圖3A係以用戶個體A為例,當用戶個體A說話時(例如念出對應的身分證字號),錄像器10擷取到用戶個體A的第一唇部動態影像,而在第一個(例如N為1)第一單位時間t1內的第一唇部動態影像如圖3A所示。運算器12在此第一個第一單位時間t1內偵測到第一特徵點x1(以上唇中央一參考點作為第一特徵點x1)由位置L1移動到位置L2,因此計算出第一距離變化量D1。The following will describe a specific implementation of how the above-mentioned
另一方面,在資料庫11中儲存有預設的第二唇部動態影像。如圖3B所示,運算器12在第一個(例如N為1)第二單位時間t2內偵測到第二唇部動態影像的第二特徵點x2 由位置L1’移動到位置L2’, 運算器12 即可計算得到第二距離變化量D2。進一步地,運算器12計算第一距離變化量D1與第二距離變化量D2的差值並且將其比對到一預設值。在此實施例中,當此差值小於預設值時,運算器12判斷第一唇部動態影像與第二唇部動態影像的相似度大於驗證閾值,此時運算器12認可用戶個體A的身分而對應產生驗證成功的訊息。反之,當此差值大於預設值時,運算器12判斷第一唇部動態影像與第二唇部動態影像的相似度小於驗證閾值,此時運算器12認定用戶個體A的身分有疑慮而對應產生驗證不成功的訊息。On the other hand, a preset second lip dynamic image is stored in the
前述實施例僅針對一單位時間內唇部動態影像的單一特徵點進行分析,然而單一特徵點可能因某些特定發音的單詞或單字而無法辨識出有效的距離變化量,導致運算器12無法準確地判斷唇部動態影像的相似度。為了因應此問題,本發明的身分驗證裝置可針對一單位時間內唇部動態影像的多個特徵點進行分析。具體來說,於一實施例中,運算器12可用以在多個第一單位時間的第N個偵測第一唇部動態影像的多個第一特徵點,以計算該些第一特徵點在第N個第一單位時間內的多個第一距離變化量,且運算器可用以在多個第二單位時間的第N個偵測第二唇部動態影像的多個第二特徵點,以計算該些第二特徵點在第N個第二單位時間內的多個第二距離變化量,其中N為正整數。接著,運算器根據該些第一距離變化量及該些第二距離變化量,據以判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值。The foregoing embodiment only analyzes a single feature point of the dynamic image of the lips within a unit of time. However, a single feature point may not be able to identify the effective distance change due to certain words or characters with specific pronunciations, causing the
以下將以第二實際範例說明上述運算器12係如何判斷第一唇部動態影像與第二唇部動態影像之間的相似度是否大於驗證閾值的具體實施方式。請進一步參照圖4A與圖4B,圖4A係依據本發明之另一實施例所繪示的用戶個體的第一唇部動態影像示意圖,而圖4B係依據本發明之另一實施例所繪示的預存之第二唇部動態影像示意圖。圖4A係以用戶個體B為例,當用戶個體B說話時(例如念出對應的身分證字號),錄像器10擷取到用戶個體B的第一唇部動態影像,而在第一個(例如N為1)第一單位時間t1內的第一唇部動態影像如圖4A所示。運算器12在此第一個第一單位時間t1內偵測第一特徵點x1及x2 (以上唇中央及側邊的參考點作為第一及第二特徵點x1及x2),其中第一特徵點x1由位置L1移動到位置L2,而第一特徵點x2由位置L3移動到位置L4,因此計算出二個第一距離變化量D1及D2。The following will describe a specific implementation of how the above-mentioned
另一方面,在資料庫11中儲存有預設的第二唇部動態影像。如圖4B所示,運算器12在第一個(例如N為1)第二單位時間t2內偵測第二唇部動態影像的第二特徵點x3及x4,其中第二特徵點x3由位置L1’移動到位置L2’,而第二特徵點x4由位置L3’移動到位置L4’。對應地,運算器12可計算得到第二距離變化量D3與D4。進一步地,運算器12計算第一距離變化量D1與第二距離變化量D3的第一差值並且將其比對到第一預設值,且運算器12計算第一距離變化量D2與第二距離變化量D4的第二差值並且將其比對到第二預設值。On the other hand, a preset second lip dynamic image is stored in the
在此實施例中,當第一與第二差值分別小於第一與第二預設值時,運算器12判斷第一唇部動態影像與第二唇部動態影像的相似度大於驗證閾值,此時運算器12認可用戶個體B的身分而對應產生驗證成功的訊息。反之,當第一差值大於第一預設值或第二差值大於第二預設值時,運算器12判斷第一唇部動態影像與第二唇部動態影像的相似度小於驗證閾值,此時運算器12認定用戶個體B的身分有疑慮而對應產生驗證不成功的訊息。In this embodiment, when the first and second differences are smaller than the first and second preset values respectively, the
前述兩個實施例僅針對一單位時間內唇部動態影像的一或多個特徵點進行分析,然而本發明的身分驗證裝置亦可針對連續的單位時間內唇部動態影像的特徵點位置變化進行分析。具體來說,於一實施例中,運算器12用以依序在每個第一單位時間內偵測第一唇部動態影像的一第一特徵點,以計算第一特徵點在第一時序內的第一移動軌跡,且運算器12更用以依序在每個第二單位時間內偵測第二唇部動態影像的第二特徵點,以計算第二特徵點在第二時序內的第二移動軌跡,運算器12根據第一移動軌跡及第二移動軌跡,據以判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值。The foregoing two embodiments only analyze one or more feature points of the dynamic image of the lips within a unit of time. However, the identity verification device of the present invention can also analyze the position changes of the feature points of the dynamic image of the lips within a continuous unit of time. analyze. Specifically, in one embodiment, the
以下將以第三實際範例說明上述運算器12係如何判斷第一唇部動態影像與第二唇部動態影像之間的相似度是否大於驗證閾值的具體實施方式。請進一步參照圖5A與圖5B,圖5A係依據本發明之另一實施例所繪示的用戶個體的第一唇部動態影像示意圖,而圖5B係依據本發明之另一實施例所繪示的預存之第二唇部動態影像示意圖。圖5A係以用戶個體C為例,為了便於說明,此實施例以三個單位時間t1~t3作為第一時序,且以三個單位時間t1’~t3’作為第二時序。當用戶個體C說話時(例如念出對應的身分證字號),錄像器10擷取到用戶個體C的第一唇部動態影像。此第一唇部動態影像在連續的不同第一單位時間t1~t3的變化如圖5A所示。運算器12依序在每個第一單位時間t1~t3內偵測第一唇部動態影像的第一特徵點,計算取得第一特徵點在第一時序(即連續的第一單位時間t1~t3)內的第一移動軌跡R1。The following will describe a specific implementation of how the above-mentioned
另一方面,在資料庫11中儲存有第二唇部動態影像。如圖5B所示,運算器12依序在每個第一單位時間t1’~t3’內偵測第二唇部動態影像的第二特徵點,計算取得第二特徵點在第二時序(即連續的第二單位時間t1’~t3’)內的第二移動軌跡R2。接著,運算器12例如利用曲線相似度演算法運算出第一移動軌跡R1與第二移動軌跡R2之間的相似度並將其對比到預設值。在此實施例中,當移動軌跡的相似度達到預設值時,運算器12判斷第一唇部動態影像與第二唇部動態影像的相似度大於驗證閾值,因此運算器12認可用戶個體C的身分而對應產生驗證成功的訊息。反之,當移動軌跡的相似度未達預設值時,運算器12判斷第一唇部動態影像與第二唇部動態影像的相似度小於驗證閾值。此時,運算器12認定用戶個體C的身分有疑慮而對應產生驗證不成功的訊息。On the other hand, the second lip dynamic image is stored in the
請參照圖6,圖6係依據本發明之一實施例所繪示的基於動態影像的身分驗證方法的方法流程圖。如圖6所示,在步驟S1中,以運算器12依據請求指令而產生控制指令。在步驟S2中,以錄像器10依據控制指令擷取包含第一語音訊息的第一唇部動態影像,其中第一語音訊息關聯於待驗證個資。在步驟S3中,以運算器12判斷待驗證個資是否符合資料庫內的預設個資。Please refer to FIG. 6 , which is a method flow chart of an identity verification method based on dynamic images according to an embodiment of the present invention. As shown in FIG. 6 , in step S1 , the
當待驗證個資符合預設個資時,在步驟S4中,以運算器12判斷第一唇部動態影像與資料庫11內的第二唇部動態影像的相似度是否大於驗證閾值,其中第二唇部動態影像包含第二語音訊息。當相似度大於驗證閾值時,在步驟S5中,以運算器12產生驗證成功訊息。When the personal information to be verified matches the preset personal information, in step S4, the
於一實施例中,第一語音訊息對應具有多個第一單位時間的第一時序,且第二語音訊息對應具有多個第二單位時間的第二時序,其中判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值包含:以運算器12在該些第一單位時間的第N個偵測第一唇部動態影像的第一特徵點,以計算第一特徵點在第N個第一單位時間內的第一距離變化量;以運算器12在該些第二單位時間的第N個偵測第二唇部動態影像的第二特徵點,以計算第二特徵點在第N個第二單位時間內的第二距離變化量,其中N為正整數;以及以運算器12運算第一距離變化量及第二距離變化量的差值。In one embodiment, the first voice message corresponds to a first time sequence having a plurality of first unit times, and the second voice message corresponds to a second time sequence having a plurality of second unit times, wherein the first lip dynamic image is determined Whether the similarity with the second lip dynamic image is greater than the verification threshold includes: using the
於一實施例中,第一語音訊息對應具有多個第一單位時間的第一時序,且第二語音訊息對應具有多個第二單位時間的第二時序,其中判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值包含以運算器12在該些第一單位時間的第N個偵測第一唇部動態影像的多個第一特徵點,以計算該些第一特徵點在第N個第一單位時間內的多個第一距離變化量;以運算器12在該些第二單位時間的第N個偵測第二唇部動態影像的多個第二特徵點,以計算該些第二特徵點在第N個第二單位時間內的多個第二距離變化量,其中N為正整數;以及以運算器12個別運算該些第一距離變化量及該些第二距離變化量的差值。In one embodiment, the first voice message corresponds to a first time sequence having a plurality of first unit times, and the second voice message corresponds to a second time sequence having a plurality of second unit times, wherein the first lip dynamic image is determined Whether the similarity to the second lip dynamic image is greater than the verification threshold includes using the
於一實施例中,第一語音訊息對應具有多個第一單位時間的第一時序,且第二語音訊息對應具有多個第二單位時間的第二時序,其中判斷第一唇部動態影像與第二唇部動態影像的相似度是否大於驗證閾值包含以運算器12依序在每個第一單位時間內偵測第一唇部動態影像的第一特徵點,以計算第一特徵點在第一時序內的第一移動軌跡;以運算器12依序在每個第二單位時間內偵測第二唇部動態影像的第二特徵點,以計算第二特徵點在第二時序內的第二移動軌跡;以及以運算器12比對第一移動軌跡及第二移動軌跡。In one embodiment, the first voice message corresponds to a first time sequence having a plurality of first unit times, and the second voice message corresponds to a second time sequence having a plurality of second unit times, wherein the first lip dynamic image is determined Whether the similarity with the second lip dynamic image is greater than the verification threshold includes using the
綜上所述,在本發明的基於動態影像的身分驗證裝置及方法中,先是利用語音訊息分析其語意內容以比對到預設個資作為第一道驗證,待通過此第一道驗證後,接著使用擷取到的唇部動態影像與預設的唇部動態影像進行相似度比對以作為第二道驗證,進而在通過兩道驗證時產生驗證成功訊息。如此一來,以雙重的語音及動態影像認證的方式重複確認用戶身份,可顯著提升用戶身份辨識的準確性且亦增加用戶的驗證過程的便利性。To sum up, in the dynamic image-based identity verification device and method of the present invention, the semantic content of the voice message is first analyzed to match the preset personal information as the first verification. After passing this first verification , and then use the captured lip dynamic image to compare the similarity with the preset lip dynamic image as the second verification, and then generate a verification success message when the two verifications are passed. In this way, repeatedly confirming the user's identity through dual voice and dynamic image authentication can significantly improve the accuracy of user identity recognition and also increase the convenience of the user's verification process.
雖然本發明以前述之實施例揭露如上,然其並非用以限定本發明。在不脫離本發明之精神和範圍內,所為之更動與潤飾,均屬本發明之專利保護範圍。關於本發明所界定之保護範圍請參考所附之申請專利範圍。Although the present invention is disclosed in the foregoing embodiments, they are not intended to limit the present invention. All changes and modifications made without departing from the spirit and scope of the present invention shall fall within the scope of patent protection of the present invention. Regarding the protection scope defined by the present invention, please refer to the attached patent application scope.
1:身分驗證裝置 10:錄像器 11:資料庫 12:運算器 DA1:第一語音訊息 DA2:第二語音訊息 d1~d5:第一音節資料 d1’~d5’:第二音節資料 t1~t5、t1’~t5’:單位時間 P1:第一時序 P2:第二時序 L1~L4、L1’~L4’:位置 x1~x4:特徵點 D1~D4:距離變化量 R1:第一移動軌跡 R2:第二移動軌跡 1: Identity verification device 10:Video recorder 11:Database 12:Operator DA1: first voice message DA2: Second voice message d1~d5: first syllable data d1’~d5’: second syllable information t1~t5, t1’~t5’: unit time P1: first timing P2: Second timing L1~L4, L1’~L4’: position x1~x4: Feature points D1~D4: distance change amount R1: first movement trajectory R2: Second movement trajectory
圖1係依據本發明之一實施例所繪示的基於動態影像的身分驗證裝置的功能方塊圖。 圖2係依據本發明之一實施例所繪示的第一語音訊息及第二語音訊息的示意圖。 圖3A係依據本發明之一實施例所繪示的用戶個體的第一唇部動態影像示意圖。 圖3B係依據本發明之一實施例所繪示的預存之第二唇部動態影像示意圖。 圖4A係依據本發明之另一實施例所繪示的用戶個體的第一唇部動態影像示意圖。 圖4B係依據本發明之另一實施例所繪示的預存之第二唇部動態影像示意圖。 圖5A係依據本發明之另一實施例所繪示的用戶個體的第一唇部動態影像示意圖。 圖5B係依據本發明之另一實施例所繪示的預存之第二唇部動態影像示意圖。 圖6係依據本發明之一實施例所繪示的基於動態影像的身分驗證方法的方法流程圖。 FIG. 1 is a functional block diagram of an identity verification device based on dynamic images according to an embodiment of the present invention. FIG. 2 is a schematic diagram of a first voice message and a second voice message according to an embodiment of the present invention. FIG. 3A is a schematic diagram of a first lip dynamic image of an individual user according to an embodiment of the present invention. FIG. 3B is a schematic diagram of a pre-stored second lip dynamic image according to an embodiment of the present invention. FIG. 4A is a schematic diagram of a first lip dynamic image of an individual user according to another embodiment of the present invention. FIG. 4B is a schematic diagram of a pre-stored second lip dynamic image according to another embodiment of the present invention. FIG. 5A is a schematic diagram of a first lip dynamic image of an individual user according to another embodiment of the present invention. FIG. 5B is a schematic diagram of a pre-stored second lip dynamic image according to another embodiment of the present invention. FIG. 6 is a method flow chart of an identity verification method based on dynamic images according to an embodiment of the present invention.
1:身分驗證裝置 1: Identity verification device
10:錄像器 10:Video recorder
11:資料庫 11:Database
12:運算器 12:Operator
Claims (8)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW112124001A TWI822646B (en) | 2019-08-07 | 2019-08-07 | Identity verification device and method based on dynamic lip image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW112124001A TWI822646B (en) | 2019-08-07 | 2019-08-07 | Identity verification device and method based on dynamic lip image |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI822646B true TWI822646B (en) | 2023-11-11 |
TW202345068A TW202345068A (en) | 2023-11-16 |
Family
ID=89720350
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW112124001A TWI822646B (en) | 2019-08-07 | 2019-08-07 | Identity verification device and method based on dynamic lip image |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI822646B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI324313B (en) * | 2006-08-25 | 2010-05-01 | Compal Electronics Inc | Identification mathod |
CN104376250A (en) * | 2014-12-03 | 2015-02-25 | 优化科技(苏州)有限公司 | Real person living body identity verification method based on sound-type image feature |
CN104657650A (en) * | 2015-01-06 | 2015-05-27 | 三星电子(中国)研发中心 | Method and device for data input or authentication |
TW201721510A (en) * | 2015-12-08 | 2017-06-16 | Trusted Solutions Corp | Compounded recognition system of face and voice print developing a more accurate, secure, and economical compounded recognition system |
TW201741921A (en) * | 2016-05-19 | 2017-12-01 | Alibaba Group Services Ltd | Identity authentication method and apparatus |
-
2019
- 2019-08-07 TW TW112124001A patent/TWI822646B/en active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI324313B (en) * | 2006-08-25 | 2010-05-01 | Compal Electronics Inc | Identification mathod |
CN104376250A (en) * | 2014-12-03 | 2015-02-25 | 优化科技(苏州)有限公司 | Real person living body identity verification method based on sound-type image feature |
CN104657650A (en) * | 2015-01-06 | 2015-05-27 | 三星电子(中国)研发中心 | Method and device for data input or authentication |
TW201721510A (en) * | 2015-12-08 | 2017-06-16 | Trusted Solutions Corp | Compounded recognition system of face and voice print developing a more accurate, secure, and economical compounded recognition system |
TW201741921A (en) * | 2016-05-19 | 2017-12-01 | Alibaba Group Services Ltd | Identity authentication method and apparatus |
Also Published As
Publication number | Publication date |
---|---|
TW202345068A (en) | 2023-11-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11126892B2 (en) | Dual stage neural network pipeline systems and methods | |
US11100481B2 (en) | Image authentication and security system and method | |
JP5285432B2 (en) | Rolling registration method and apparatus for signature verification | |
US20190258886A1 (en) | Extraction of data from a digital image | |
US20170011405A1 (en) | Simultaneous multi-factor authentication systems and methods for payment transactions | |
US10019641B2 (en) | Background OCR during card data entry | |
JP2018520397A (en) | Method and system for identifying human or machine | |
WO2019130670A1 (en) | Biometric authentication system | |
US10878071B2 (en) | Biometric authentication anomaly detection | |
US10713466B2 (en) | Fingerprint recognition method and electronic device using the same | |
US10089349B2 (en) | Method and electronic device for updating the registered fingerprint datasets of fingerprint recognition | |
US20190303659A1 (en) | User identification system and method for identifying user | |
KR101535591B1 (en) | Authentication processing method, authentication processing system performing the same, c0mputer program for the same and storage medium storing the same | |
US20220245969A1 (en) | Apparatus and method for user authentication based on face recognition and handwritten signature verification | |
JP4665764B2 (en) | Pattern identification system, pattern identification method, and pattern identification program | |
US20150254498A1 (en) | Fingerprint recognition methods and devices | |
TWI822646B (en) | Identity verification device and method based on dynamic lip image | |
US20070233667A1 (en) | Method and apparatus for sample categorization | |
TWI801647B (en) | Identity verification device and method based on dynamic image | |
US9613252B1 (en) | Fingerprint matching method and device | |
KR20230156823A (en) | Fingerprint identification method, device, electronic apparatus and storage medium | |
TWM588311U (en) | Identity verification device based on dynamic image | |
TWM626411U (en) | Cardless finance transaction system and its host server | |
TWM555521U (en) | Hospitality service system | |
KR100944248B1 (en) | System and method for user recognition |