TW202141310A - Biometric data encryption device and method and information processing device using the method wherein the entire process of biometric matching and identity recognition does not expose the biometric image and successfully prevents theft of the user's biometric image data, thereby enhancing the high degree of protection of the user's biometrics - Google Patents
Biometric data encryption device and method and information processing device using the method wherein the entire process of biometric matching and identity recognition does not expose the biometric image and successfully prevents theft of the user's biometric image data, thereby enhancing the high degree of protection of the user's biometrics Download PDFInfo
- Publication number
- TW202141310A TW202141310A TW109114234A TW109114234A TW202141310A TW 202141310 A TW202141310 A TW 202141310A TW 109114234 A TW109114234 A TW 109114234A TW 109114234 A TW109114234 A TW 109114234A TW 202141310 A TW202141310 A TW 202141310A
- Authority
- TW
- Taiwan
- Prior art keywords
- biometric
- image
- data
- encrypted
- unit
- Prior art date
Links
Images
Abstract
Description
本發明係關於生物特徵採集與加密技術之相關領域,尤指一種生物特徵資料加密裝置與方法。The present invention relates to the related fields of biometric collection and encryption technology, in particular to a biometric data encryption device and method.
隨著各種智慧終端機的發展,觸控顯示屏幕成為主要的人機介面。進一步地,隨著全屏幕智能手機逐漸成為主流,屏下式光學式生物特徵(指紋、掌紋)辨識裝置也已經廣泛地整合在全屏幕智能手機之中。生物特徵具有唯一性、隱私性和不可更改性等特性,因此,各式網路金融服務無不透過生物特徵相互匹配的方式,完成每位個體客戶的身份識別。With the development of various smart terminals, touch-sensitive display screens have become the main human-machine interface. Furthermore, as full-screen smart phones gradually become the mainstream, under-screen optical biometric (fingerprints, palmprints) recognition devices have also been widely integrated into full-screen smart phones. Biometrics have the characteristics of uniqueness, privacy, and immutability. Therefore, all kinds of online financial services can complete the identification of each individual customer through the matching of biometrics.
圖1顯示習知的一種生物特徵採集與識別裝置的方塊圖。如圖1所示,習知的生物特徵(指紋)採集與識別裝置包含一光檢測器電路2’和一生物特徵檢測電路1’,其中該光檢測器電路2’通常包含複數個光二極體感測單元21’,且其整合在智能手機3’的觸控顯示屏幕31’的下方處。另一方面,該生物特徵檢測電路1’包含一指紋資料採集模塊11’和一應用處理器12’,其中該指紋資料採集模塊11’至少包含一電荷採樣單元、一信號放大單元、以及一類比數位轉換單元。Figure 1 shows a block diagram of a conventional biological feature collection and recognition device. As shown in Figure 1, the conventional biometric (fingerprint) collection and identification device includes a photodetector circuit 2'and a biometric detection circuit 1', wherein the photodetector circuit 2'usually includes a plurality of photodiodes. The sensing unit 21' is integrated under the touch display screen 31' of the smart phone 3'. On the other hand, the biometric detection circuit 1'includes a fingerprint data collection module 11' and an application processor 12', wherein the fingerprint data collection module 11' includes at least a charge sampling unit, a signal amplification unit, and an analog Digital conversion unit.
更詳細地說明,所述應用處理器12’包含一資料處理單元121’、一儲存單元122’以及一識別單元123’。建立一指紋圖像模板時,該資料處理單元121’將該指紋資料採集模塊11’所傳送的指紋採集資料進一步處理成一指紋圖像,而後將該指紋圖像儲存在該儲存單元122’之中以作為一指紋圖像模板。進行指紋識別程序時,該資料處理單元121’自該指紋資料採集模塊11’所傳送的指紋採集資料之中提取出複數個指紋特徵,接著該識別單元123’對該複數個指紋特徵和該指紋圖像模板進行一特徵匹配。In more detail, the application processor 12' includes a data processing unit 121', a storage unit 122', and an identification unit 123'. When creating a fingerprint image template, the data processing unit 121' further processes the fingerprint collection data sent by the fingerprint data collection module 11' into a fingerprint image, and then stores the fingerprint image in the storage unit 122' Take it as a fingerprint image template. When performing a fingerprint identification procedure, the data processing unit 121' extracts a plurality of fingerprint features from the fingerprint collection data sent by the fingerprint data collection module 11', and then the identification unit 123' extracts the plurality of fingerprint features and the fingerprint The image template performs a feature matching.
由前述說明可知,習知技術通常將用戶的指紋圖像模板儲存於儲存單元122’,然而,這種方式存在許多安全漏洞,從而導致用戶的指紋圖像模板有外洩的風險。舉例而言,駭客可以利用一惡意程式攔截指紋資料採集模塊11’所傳送的指紋採集資料或者直接自該儲存單元122’之中複製所述指紋圖像模板。可想而知,在獲取特定個體用戶的指紋採集資料及/或紋圖像模板之後,便可以恣意地使用該特定個體用戶的網路金融服務,例如:行動支付、第三方支付、和網路銀行等。更進一步地,若用戶的指紋資料被進一步地使用在移民身分驗證、犯罪紀錄等領域的時候,則嚴重性便又提升至更高的層級。As can be seen from the foregoing description, the conventional technology usually stores the user's fingerprint image template in the storage unit 122'. However, this method has many security loopholes, which may cause the user's fingerprint image template to leak. For example, a hacker can use a malicious program to intercept the fingerprint collection data sent by the fingerprint data collection module 11' or directly copy the fingerprint image template from the storage unit 122'. It is conceivable that after obtaining fingerprint collection data and/or pattern image templates of a specific individual user, you can freely use the online financial services of that specific individual user, such as: mobile payment, third-party payment, and Internet Bank etc. Furthermore, if the user's fingerprint data is further used in areas such as immigration identity verification, criminal records, etc., the severity will be raised to a higher level.
由上述說明可知,習知的生物特徵(指紋)採集與識別裝置在使用安全性方面仍具有顯著的實務運作缺陷,因此本領域亟需一種新式的生物特徵資料加密裝置與方法。From the above description, it can be seen that the conventional biometric (fingerprint) collection and identification device still has significant practical operational defects in terms of use security. Therefore, there is an urgent need for a new type of biometric data encryption device and method in this field.
本發明之主要目的在於提供一種生物特徵資料加密裝置及方法,其應用於一生物特徵採集與辨識電路之中,使該生物特徵採集與辨識電路在獲得一生物特徵(指紋)圖像之後,能夠接著將該生物特徵圖像進行加密處理。並且,在利用生物特徵匹配執行身分識別時,需利用卷積神經網路演算法對加密後的生物特徵數據進行特徵提取,從而使得特徵匹配及身分識別的全部過程皆不會暴露生物特徵圖像,成功地對用戶的生物特徵(例如:指紋)圖像數據起到了防止竊取的功效,從而提升用戶生物特徵之高度保護性。The main purpose of the present invention is to provide a biometric data encryption device and method, which are applied to a biometrics acquisition and identification circuit, so that the biometrics acquisition and identification circuit can obtain a biometric (fingerprint) image. Then the biometric image is encrypted. In addition, when using biometric matching to perform identity recognition, it is necessary to use a convolutional neural network algorithm to perform feature extraction on the encrypted biometric data, so that the entire process of feature matching and identity recognition does not expose the biometric image. It has successfully prevented the theft of the user's biometric (for example: fingerprint) image data, thereby enhancing the high degree of protection of the user's biometrics.
為達成上述目的,本發明提出所述生物特徵資料加密裝置之一實施例,其應用於一生物特徵採集與辨識電路之中,其中該生物特徵採集與辨識電路包含一生物特徵資料採集模塊和一應用處理器,且該應用處理器具有一資料處理單元和一儲存單元;該生物特徵資料加密裝置包括:In order to achieve the above objective, the present invention proposes an embodiment of the biometric data encryption device, which is applied to a biometric data collection and identification circuit, wherein the biometric data collection and identification circuit includes a biometric data collection module and a biometric data collection module. An application processor, and the application processor has a data processing unit and a storage unit; the biometric data encryption device includes:
一加密單元,整合在該應用處理器之中;在該資料處理單元將該生物特徵資料採集模塊所傳送的一生物特徵採集資料進一步處理成一生物特徵圖像之後,該加密單元依序對該生物特徵圖像執行一圖像邊界擴充處理、一圖像分割處理和一圖像像素矩陣生成處理以獲得X×Y個圖像像素矩陣;在依據所述X×Y個圖像像素矩陣進一步獲得用以組成一加密數據的m×n個數據之後,該加密單元將該加密數據和該生物特徵圖像整合為一加密生物特徵數據,且將該加密生物特徵數據儲存在該儲存單元之中;其中,X、Y、m、n皆為正整數,且m和n皆可以被8整除。An encryption unit is integrated in the application processor; after the data processing unit further processes a biometric data collected from the biometric data collection module into a biometric image, the encryption unit sequentially processes the biometric image. The feature image performs an image boundary expansion process, an image segmentation process, and an image pixel matrix generation process to obtain X×Y image pixel matrices; based on the X×Y image pixel matrices, the use of After forming an encrypted data with m×n data, the encryption unit integrates the encrypted data and the biometric image into an encrypted biometric data, and stores the encrypted biometric data in the storage unit; wherein , X, Y, m, n are all positive integers, and m and n are all divisible by 8.
在一實施例中,前述本發明之生物特徵資料加密裝置更包括:In one embodiment, the aforementioned biometric data encryption device of the present invention further includes:
一特徵比對單元,整合在該應用處理器之中,用以存取該儲存單元以獲得所述加密生物特徵數據;在需要進行一特徵比對程序的情況下,該特徵比對單元對該加密生物特徵數據和一待比對生物特徵圖像分別執行一特徵提取處理以獲得一加密生物特徵和一待比對生物特徵,從而進一步比對該待比對生物特徵是否與該加密生物特徵相互匹配。A feature comparison unit integrated in the application processor for accessing the storage unit to obtain the encrypted biometric data; in the case of a feature comparison program, the feature comparison unit The encrypted biometric data and a biometric image to be compared perform a feature extraction process respectively to obtain an encrypted biometric and a biometric to be compared, so as to further compare whether the biometric to be compared and the encrypted biometric are mutually. match.
在一實施例中,前述本發明之生物特徵資料加密裝置更包括:In one embodiment, the aforementioned biometric data encryption device of the present invention further includes:
一解密單元,整合在該應用處理器之中;其中,在需要對所述加密生物特徵數據執行一解密程序以獲得所述生物特徵圖像的情況下,該解密單元存取該儲存單元以獲得該加密生物特徵數據,且將該加密生物特徵數據分成m×n/X×Y個子數據,從而依據所述m×n/X×Y個子數據生成m×n/X×Y個圖像區塊,進以將所述m×n/X×Y個圖像區塊組成所述生物特徵圖像。A decryption unit integrated in the application processor; wherein, when a decryption program needs to be performed on the encrypted biometric data to obtain the biometric image, the decryption unit accesses the storage unit to obtain The encrypted biometric data, and the encrypted biometric data is divided into m×n/X×Y sub-data, thereby generating m×n/X×Y image blocks according to the m×n/X×Y sub-data , And further compose the m×n/X×Y image blocks into the biometric image.
在一實施例中,完成所述圖像邊界擴充處理的該生物特徵圖像具有m×n的一圖像尺寸,且包含複數個擴充像素,各所述擴充像素的一灰階值皆為0。In one embodiment, the biometric image that has completed the image boundary expansion processing has an image size of m×n and includes a plurality of expanded pixels, and each of the expanded pixels has a grayscale value of 0. .
在一實施例中,m值與n值相等,使得各所述圖像像素矩陣成為一圖像像素方形矩陣。In an embodiment, the value of m is equal to the value of n, so that each of the image pixel matrix becomes a square matrix of image pixels.
在一實施例中,該加密單元具有:In an embodiment, the encryption unit has:
一圖像邊界擴充單元,用以對該生物特徵圖像執行所述圖像邊界擴充處理;An image boundary expansion unit for performing the image boundary expansion processing on the biometric image;
一圖像分割單元,用以對完成所述圖像邊界擴充處理的該生物特徵圖像執行所述圖像分割處理,從而獲得m×n/X×Y個圖像區塊;An image segmentation unit for performing the image segmentation process on the biometric image that has completed the image boundary expansion process, so as to obtain m×n/X×Y image blocks;
一矩陣生成單元,依據所述m×n/X×Y個圖像區塊生成所述X×Y個圖像像素矩陣,且將各所述圖像像素矩陣所含有的-1置換成0;A matrix generating unit, which generates the X×Y image pixel matrixes according to the m×n/X×Y image blocks, and replaces -1 contained in each of the image pixel matrices with 0;
一加密數據產生單元,用以將所述X×Y個圖像像素矩陣執行一矩陣排列處理之後,該加密數據產生單元進一步將所述m×n/X×Y個圖像區塊和所述X×Y個圖像像素矩陣進行一對應位置相乘,從而獲得用以組成所述加密數據的m×n個所述數據,進以將該加密數據和該生物特徵圖像整合為所述加密生物特徵數據。An encrypted data generating unit for performing a matrix arrangement process on the X×Y image pixel matrix, the encrypted data generating unit further combines the m×n/X×Y image blocks with the X×Y image pixel matrices are multiplied by a corresponding position to obtain m×n pieces of the data used to form the encrypted data, and then the encrypted data and the biometric image are integrated into the encrypted data. Biometric data.
在一實施例中,該解密單元具有:In an embodiment, the decryption unit has:
一數據分割單元,用以存取該儲存單元以獲得該加密生物特徵數據,且將該加密生物特徵數據分成m×n/X×Y個所述子數據;A data dividing unit for accessing the storage unit to obtain the encrypted biometric data, and dividing the encrypted biometric data into m×n/X×Y sub-data;
一圖像區塊重建單元,用以依據所述m×n/X×Y個子數據生成m×n/X×Y個所述圖像區塊;以及An image block reconstruction unit for generating m×n/X×Y image blocks according to the m×n/X×Y sub-data; and
一圖像重建單元,用以將所述m×n/X×Y個組成所述生物特徵圖像。An image reconstruction unit for composing the m×n/X×Y into the biometric image.
為達成上述目的,本發明同時提出所述生物特徵資料加密方法之一實施例,其應用於一生物特徵採集與辨識電路之中,其中該生物特徵採集與辨識電路包含一生物特徵資料採集模塊和一應用處理器,且該生物特徵資料加密方法包括:To achieve the above objective, the present invention also proposes an embodiment of the biometric data encryption method, which is applied to a biometric collection and identification circuit, wherein the biometric collection and identification circuit includes a biometric data collection module and An application processor, and the biometric data encryption method includes:
一加密程序,具有以下步驟:An encryption procedure with the following steps:
在該應用處理器的一資料處理單元將一生物特徵採集資料處理成一生物特徵圖像之後,依序對該生物特徵圖像執行一圖像邊界擴充處理、一圖像分割處理和一圖像像素矩陣生成處理以獲得X×Y個圖像像素矩陣;其中,X和Y皆為正整數;及After a data processing unit of the application processor processes a biometric feature collection data into a biometric image, an image boundary expansion process, an image segmentation process, and an image pixel are sequentially performed on the biometric image Matrix generation processing to obtain X×Y image pixel matrix; where X and Y are both positive integers; and
依據所述X×Y個圖像像素矩陣進一步獲得用以組成一加密數據的m×n個數據,將該加密數據和該生物特徵圖像整合為一加密生物特徵數據,且將該加密生物特徵數據儲存在該應用處理器的一儲存單元之中;其中, m和n皆為正整數,且皆可以被8整除。According to the X×Y image pixel matrix, m×n data used to form an encrypted data are further obtained, the encrypted data and the biometric image are integrated into an encrypted biometric data, and the encrypted biometrics The data is stored in a storage unit of the application processor; where m and n are both positive integers, and both are divisible by 8.
在一實施例中,前述本發明之生物特徵資料加密方法,更包括一特徵比對程序,其具有以下步驟:In one embodiment, the aforementioned biometric data encryption method of the present invention further includes a feature comparison program, which has the following steps:
存取該儲存單元以獲得所述加密生物特徵數據,且對該加密生物特徵數據和一待比對生物特徵圖像分別執行一特徵提取處理以獲得一加密生物特徵和一待比對生物特徵;及Access the storage unit to obtain the encrypted biometric data, and perform a feature extraction process on the encrypted biometric data and a biometric image to be compared to obtain an encrypted biometric and a biometric to be compared; and
比對該待比對生物特徵是否與該加密生物特徵相互匹配。Compare whether the biological characteristics to be compared and the encrypted biological characteristics match each other.
在一實施例中,前述本發明之生物特徵資料加密方法,更包括一解密程序,其具有以下步驟:In one embodiment, the aforementioned biometric data encryption method of the present invention further includes a decryption program, which has the following steps:
存取該儲存單元以獲得該加密生物特徵數據,且將該加密生物特徵數據分成m×n/X×Y個子數據,從而依據所述m×n/X×Y個子數據生成m×n/X×Y個圖像區塊,進以將所述m×n/X×Y個圖像區塊組成所述生物特徵圖像。Access the storage unit to obtain the encrypted biometric data, and divide the encrypted biometric data into m×n/X×Y sub-data, thereby generating m×n/X according to the m×n/X×Y sub-data ×Y image blocks, and then the m×n/X×Y image blocks are formed into the biometric image.
本發明同時提供一種資訊處理裝置,其具有一生物特徵檢測裝置,該生物特徵檢測裝置具有一生物特徵採集與辨識電路;其中,該生物特徵採集與辨識電路包含一生物特徵資料採集模塊和一應用處理器,且該應用處理器具有一資料處理單元、一儲存單元以及如前所述本發明之生物特徵資料加密裝置。The present invention also provides an information processing device, which has a biological feature detection device, the biological feature detection device has a biological feature collection and identification circuit; wherein, the biological feature collection and identification circuit includes a biological feature data collection module and an application Processor, and the application processor has a data processing unit, a storage unit, and the biometric data encryption device of the present invention as described above.
本發明同時提供一種資訊處理裝置,其具有一生物特徵檢測採集裝置,該生物特徵檢測裝置具有一生物特徵採集與辨識電路;其中,該生物特徵採集與辨識電路包含一生物特徵資料採集模塊和一應用處理器,且該應用處理器將該生物特徵資料採集模塊所傳送的一生物特徵採集資料處理成一生物特徵圖像,且接著執行如前所述本發明之生物特徵資料加密方法。The present invention also provides an information processing device, which has a biological feature detection and collection device, the biological feature detection device has a biological feature collection and identification circuit; wherein the biological feature collection and identification circuit includes a biological feature data collection module and a The application processor, and the application processor processes a biometric data collected from the biometric data collection module into a biometric image, and then executes the biometric data encryption method of the present invention as described above.
在可行的實施例中,所述資訊處理裝置可為智能手機、平板電腦、筆記型電腦、一體式電腦、智能手錶或門禁裝置。In a feasible embodiment, the information processing device may be a smart phone, a tablet computer, a notebook computer, an all-in-one computer, a smart watch, or an access control device.
為使 貴審查委員能進一步瞭解本發明之結構、特徵、目的、與其優點,茲附以圖式及較佳具體實施例之詳細說明如後。In order to enable your reviewer to further understand the structure, features, purpose, and advantages of the present invention, the drawings and detailed descriptions of preferred specific embodiments are attached as follows.
圖2顯示包含本發明之一種生物特徵資料加密裝置的一生物特徵採集與辨識裝置的方塊圖。如圖2所示,生物特徵採集與識別裝置包含一檢測器電路2和一生物特徵採集與辨識電路1。舉例而言,該檢測器電路2為包含複數個光二極體感測單元21的光檢測器電路2,但不以此為限。各所述光二極體感測單元21整合在智能手機3的觸控顯示屏幕31的下方處。另一方面,該生物特徵採集與辨識電路1包含一生物特徵資料採集模塊11和一應用處理器12。舉例而言,若所述該生物特徵採集與辨識電路1為一指紋採集與辨識電路,則該生物特徵資料採集模塊11基礎上包含:一電荷採樣單元、一信號放大單元、以及一類比數位轉換單元。並且,該應用處理器12整合有一資料處理單元121、一儲存單元122以及本發明之一種生物特徵資料加密裝置。如圖2所示,本發明之生物特徵資料加密裝置包含一加密單元123、一特徵比對單元124以及一解密單元125。Fig. 2 shows a block diagram of a biological feature collection and identification device including a biological feature data encryption device of the present invention. As shown in FIG. 2, the biological feature collection and identification device includes a
本發明同時提供一種生物特徵資料加密方法,且圖3顯示本發明之一種生物特徵資料加密方法的流程圖。在建立一指紋圖像模板時,該資料處理單元121將該指紋資料採集模塊11所傳送的一指紋採集資料進一步處理成一指紋圖像,而後將該指紋圖像儲存在該儲存單元122之中以作為一指紋圖像模板。此時,本發明之生物特徵資料加密裝置即開始運作,從而以其所述加密單元123依序對該生物特徵圖像執行一圖像邊界擴充處理、一圖像分割處理和一圖像像素矩陣生成處理以獲得X×Y個圖像像素矩陣(亦即,執行步驟S1)。其中,X和Y皆為正整數。The present invention also provides a method for encrypting biometric data, and FIG. 3 shows a flowchart of the method for encrypting biometric data of the present invention. When creating a fingerprint image template, the
圖4顯示本發明之生物特徵資料加密裝置的加密單元的方塊圖。如圖2、圖3與圖4所示,執行步驟S1時,該加密單元123以其一圖像邊界擴充單元1231對由儲存在該儲存單元122之中的生物特徵圖像執行一圖像邊界擴充處理,使得完成所述圖像邊界擴充處理的該生物特徵圖像具有m×n的一圖像尺寸。舉例而言,生物特徵圖像的原圖像尺寸為198×198,經過所述圖像邊界擴充處理之後,生物特徵圖像具有200×200的圖像尺寸。亦即,經過所述圖像邊界擴充處理之後,m(圖像寬度)和n(圖像長度)皆為可以被8整除的正整數。補充說明的是,經過所述圖像邊界擴充處理之後,生物特徵圖像會包含複數個擴充像素,其中各所述擴充像素的灰階值皆為0。Figure 4 shows a block diagram of the encryption unit of the biometric data encryption device of the present invention. As shown in FIGS. 2, 3, and 4, when step S1 is performed, the
接著,該加密單元1231以其一圖像分割單元1232對完成所述圖像邊界擴充處理的該生物特徵圖像執行一圖像分割處理,從而獲得m×n/X×Y個圖像區塊。舉例而言,基於8×8的基礎區塊將該生物特徵圖像分割成m×n/64個圖像區塊。繼續地,該加密單元1231以其一矩陣生成單元1233依據所述m×n/X×Y個圖像區塊生成X×Y個圖像像素矩陣。例如,生成64個圖像像素矩陣,且各所述圖像像素矩陣皆為一8階阿達瑪(Hadamard)矩陣。圖5即顯示8階阿達瑪矩陣的一示圖。進一步地,圖6顯示8階阿達瑪矩陣的另一示圖。繼續地,該加密單元1231以其一加密數據產生單元1234以將各所述圖像像素矩陣所含有的-1置換成0。更詳細地說明示,所述加密數據產生單元1234將圖5所示的8階阿達瑪矩陣的一示圖所含有的-1置換成0之後,即獲得圖6所示的8階阿達瑪矩陣。Next, the
請重複參閱圖2、圖3與圖4,獲得X×Y個(例如:64個)圖像像素矩陣之後,方法流程即執行步驟S2,從而令加密單元123以其一加密數據產生單元1234將所述X×Y個圖像像素矩陣執行一矩陣排列處理,且進一步將所述m×n/X×Y個(例如64個)圖像區塊和所述X×Y個(例如64個)圖像像素矩陣進行一對應位置相乘,藉此方式獲得用以組成一加密數據的m×n個數據。之後,該加密單元123將該加密數據和該生物特徵圖像整合為一加密生物特徵數據。舉例而言,加密單元123將64個8階阿達瑪(Hadamard)矩陣進行排序,獲得排序後的64個8階阿達瑪(Hadamard)矩陣{H1
, H2
,……,H64
},並將此一排序方式K予以記錄。接著,將具有8×8的圖像尺寸的各所述圖像區塊依照對應位置與各所述8階阿達瑪矩陣相乘,從而獲得64個(亦即,m×n個)數據。如此,這64個數據便可組合成一加密數據,最終與該生物特徵圖像整合為一加密生物特徵數據。Please refer to Figure 2, Figure 3 and Figure 4 repeatedly, after obtaining X×Y (for example: 64) image pixel matrices, the method flow is to execute step S2, so that the
值得說明的是,前述步驟S1和步驟S2為本發明之生物特徵資料加密方法的加密程序。並且,完成所述加密程序之後,本發明之生物特徵資料加密方法便可接著執行一特徵比對程序。如圖2與圖3所示,進行特徵比對程序時,該特徵比對單元124存取該儲存單元122以獲得所述加密生物特徵數據(亦即,執行步驟S3),且對該加密生物特徵數據和一待比對生物特徵圖像分別執行一特徵提取處理以獲得一加密生物特徵和一待比對生物特徵,從而進一步比對該待比對生物特徵是否與該加密生物特徵相互匹配。It is worth noting that the aforementioned steps S1 and S2 are the encryption procedures of the biometric data encryption method of the present invention. Moreover, after the encryption procedure is completed, the biometric data encryption method of the present invention can then execute a feature comparison procedure. As shown in FIG. 2 and FIG. 3, when performing the feature comparison procedure, the
本發明並不限定實現特徵提取處理時所採用的演算法之種類,在可行的實施例中,可採用卷積神經網路(CNN)演算法,例如:全卷積網路演算法(Fully convolutional neural network, FCN)、基於區域的卷積網路演算法(Region-based convolutional neural network, R-CNN)、快速的基於區域的卷積網路演算法(Fast region-based convolutional neural network, fast R-CNN)、或加快速的基於區域的卷積網路演算法(Faster region-based convolutional neural network, faster R-CNN)、或使用遮罩的基於區域的卷積網路演算法(Mask R-CNN)。The present invention does not limit the type of algorithm used in the feature extraction process. In a feasible embodiment, a convolutional neural network (CNN) algorithm can be used, for example: Fully convolutional neural network algorithm (Fully convolutional neural network) network, FCN), region-based convolutional neural network (R-CNN), fast region-based convolutional neural network (Fast region-based convolutional neural network, fast R-CNN) , Or add fast region-based convolutional neural network (Faster region-based convolutional neural network, faster R-CNN), or use masking region-based convolutional network algorithm (Mask R-CNN).
在使用卷積神經網路(CNN)演算法對該加密生物特徵數據完成所述特徵提取處理,所獲得的生物特徵數據可能包括&*#&($#(&*@#()()*%$()$%(*)$)%(。簡單地說,在利用加密單元123將加密數據和生物特徵圖像整合為所述加密生物特徵數據之後,利用特徵提取處理所提取出的生物特徵數據會顯示為亂碼,顯示本發明之生物特徵資料加密裝置成功地對用戶的生物特徵(例如:指紋)圖像數據起到了防止竊取的功效。After using the Convolutional Neural Network (CNN) algorithm to complete the feature extraction process for the encrypted biometric data, the obtained biometric data may include &*#&($#(&*@#()()* %$()$%(*)$)%(. Simply put, after using the
當然,本發明之生物特徵資料加密裝置也能夠將所述加密生物特徵數據恢復(還原)成該生物特徵圖像的方法。如圖2所示,本發明之生物特徵資料加密裝置還具有一解密單元124,其整合在該應用處理器12之中。圖7顯示本發明之生物特徵資料加密裝置的解密單元的方塊圖。執行解密程序時,該解密單元124以其一數據分割單元1251存取該儲存單元122以獲得所述加密生物特徵數據,且將該加密生物特徵數據分成m×n/X×Y個子數據。進一步地,該解密單元124以其一圖像區塊重建單元1252依據所述m×n/X×Y個子數據生成m×n/X×Y個所述圖像區塊。最終,該解密單元124以其一圖像重建單元1253將所述m×n/X×Y個圖像區塊拼接所述生物特徵圖像。Of course, the biometric data encryption device of the present invention can also restore (restore) the encrypted biometric data into the biometric image. As shown in FIG. 2, the biometric data encryption device of the present invention also has a
舉例而言,將加密生物特徵數據分成m×n/64個子數據,其中各所述子數據包含64個數據,記為{S1 , S2 ,……,S64 }。接著,基於特別紀錄的排序方式K,利用下式(1)將64個數據還原成64個圖像區塊:G=<Hi Si > - <Hi ><Si >………….(1)。For example, the encrypted biometric data is divided into m×n/64 sub-data, where each of the sub-data contains 64 data, denoted as {S 1 , S 2 ,..., S 64 }. Then, based on the sorting method K of the special record, use the following formula (1) to restore 64 data into 64 image blocks: G=<H i S i >-<H i ><S i >………… .(1).
獲得64個圖像區塊之後,所述圖像重建單元1253即將所述m×n/X×Y個組成所述生物特徵圖像。After obtaining the 64 image blocks, the
如此,上述已完整且清楚地說明本發明之一種生物特徵資料加密裝置及方法;並且,經由上述可得知本發明具有下列優點:In this way, the above has completely and clearly explained a biometric data encryption device and method of the present invention; and, from the above, it can be seen that the present invention has the following advantages:
(1)本發明所揭示之生物特徵資料加密裝置及方法,其應用於一生物特徵採集與辨識電路之中,使該生物特徵採集與辨識電路在獲得一生物特徵(指紋)圖像之後,能夠接著將該生物特徵圖像進行加密處理。並且,在利用生物特徵匹配執行身分識別時,需利用卷積神經網路演算法對加密後的生物特徵數據進行特徵提取,從而使得特徵匹配及身分識別的全部過程皆不會暴露生物特徵圖像,成功地對用戶的生物特徵(例如:指紋)圖像數據起到了防止竊取的功效,從而提升用戶生物特徵之高度保護性。(1) The biological feature data encryption device and method disclosed in the present invention are applied to a biological feature collection and identification circuit, so that the biological feature collection and identification circuit can obtain a biological feature (fingerprint) image. Then the biometric image is encrypted. In addition, when using biometric matching to perform identity recognition, it is necessary to use a convolutional neural network algorithm to perform feature extraction on the encrypted biometric data, so that the entire process of feature matching and identity recognition does not expose the biometric image. It has successfully prevented the theft of the user's biometric (for example: fingerprint) image data, thereby enhancing the high degree of protection of the user's biometrics.
(2)並且,本發明同時揭示一種資訊處理裝置,其具有一生物特徵採集與識別裝置,其具有一生物特徵採集與辨識電路;其中,該生物特徵採集與辨識電路包含一生物特徵資料採集模塊和一應用處理器,且該應用處理器具有一資料處理單元、一儲存單元以及如前所述本發明之生物特徵資料加密裝置。(2) Moreover, the present invention also discloses an information processing device, which has a biological feature collection and identification device, and has a biological feature collection and identification circuit; wherein the biological feature collection and identification circuit includes a biological feature data collection module And an application processor, and the application processor has a data processing unit, a storage unit, and the biometric data encryption device of the present invention as described above.
(3)本發明同時揭示另一種資訊處理裝置,其具有一生物特徵採集與識別裝置,該生物特徵採集與識別裝置具有一生物特徵採集與辨識電路;其中,該生物特徵採集與辨識電路包含一生物特徵資料採集模塊和一應用處理器,且該應用處理器將該生物特徵資料採集模塊所傳送的一生物特徵採集資料處理成一生物特徵圖像,且接著執行如前所述本發明之生物特徵資料加密方法。(3) The present invention also discloses another information processing device, which has a biological feature collection and identification device, the biological feature collection and identification device has a biological feature collection and identification circuit; wherein, the biological feature collection and identification circuit includes a The biometric data collection module and an application processor, and the application processor processes a biometric data collected from the biometric data collection module into a biometric image, and then executes the biometric image of the present invention as described above Data encryption method.
(4)在可行的實施例中,所述資訊處理裝置可以是智能手機、平板電腦、筆記型電腦、一體式電腦、智能手錶、或門禁裝置。(4) In a feasible embodiment, the information processing device may be a smart phone, a tablet computer, a notebook computer, an all-in-one computer, a smart watch, or an access control device.
必須加以強調的是,前述本案所揭示者乃為較佳實施例,舉凡局部之變更或修飾而源於本案之技術思想而為熟習該項技藝之人所易於推知者,俱不脫本案之專利權範疇。It must be emphasized that the foregoing disclosure in this case is a preferred embodiment, and any partial changes or modifications that are derived from the technical ideas of this case and are easily inferred by those who are familiar with the art will not deviate from the patent of this case. Right category.
綜上所陳,本案無論目的、手段與功效,皆顯示其迥異於習知技術,且其首先發明合於實用,確實符合發明之專利要件,懇請 貴審查委員明察,並早日賜予專利俾嘉惠社會,是為至禱。In summary, regardless of the purpose, means and effects of this case, it is shown that it is very different from the conventional technology, and its first invention is suitable for practicality, and it does meet the patent requirements of the invention. Please check it out and grant the patent as soon as possible. Society is for the best prayer.
<本發明> 1:生物特徵採集與辨識電路 11:生物特徵資料採集模塊 12:應用處理器 121:資料處理單元 122:儲存單元 123:加密單元 1231:圖像邊界擴充單元 1232:圖像分割單元 1233:矩陣生成單元 1234:加密數據產生單元 124:特徵比對單元 125:解密單元 1251:數據分割單元 1252:圖像區塊重建單元 1253:圖像重建單元 2:檢測器電路 21:光二極體感測單元 3:智能手機 31:觸控顯示屏幕 S1步驟:在該應用處理器的一資料處理單元將一生物特徵採集資料處理成一生物特徵圖像之後,依序對該生物特徵圖像執行一圖像邊界擴充處理、一圖像分割處理和一圖像像素矩陣生成處理以獲得X×Y個圖像像素矩陣 S2步驟:依據所述X×Y個圖像像素矩陣進一步獲得用以組成一加密數據的m×n個數據,將該加密數據和該生物特徵圖像整合為一加密生物特徵數據,且將該加密生物特徵數據儲存在該應用處理器的一儲存單元之中 S3步驟:存取該儲存單元以獲得所述加密生物特徵數據,且對該加密生物特徵數據和一待比對生物特徵圖像分別執行一特徵提取處理以獲得一加密生物特徵和一待比對生物特徵 S4步驟:比對該待比對生物特徵是否與該加密生物特徵相互匹配<The present invention> 1: Biometrics acquisition and identification circuit 11: Biometric data collection module 12: Application processor 121: Data Processing Unit 122: storage unit 123: encryption unit 1231: Image boundary expansion unit 1232: Image segmentation unit 1233: Matrix generation unit 1234: Encrypted data generation unit 124: Feature comparison unit 125: Decryption Unit 1251: data segmentation unit 1252: Image block reconstruction unit 1253: Image reconstruction unit 2: Detector circuit 21: Optical diode sensing unit 3: smart phone 31: Touch display screen Step S1: After a data processing unit of the application processor processes a biometric feature collection data into a biometric image, sequentially perform an image boundary expansion process, an image segmentation process, and a biometric image on the biometric image. Image pixel matrix generation and processing to obtain X×Y image pixel matrix Step S2: According to the X×Y image pixel matrix, m×n data used to form an encrypted data are further obtained, the encrypted data and the biometric image are integrated into an encrypted biometric data, and the The encrypted biometric data is stored in a storage unit of the application processor Step S3: Access the storage unit to obtain the encrypted biometric data, and perform a feature extraction process on the encrypted biometric data and a biometric image to be compared to obtain an encrypted biometric and a biometric image to be compared Biological characteristics Step S4: Compare whether the biometric to be compared and the encrypted biometric match each other
<習知> 1’ 生物特徵檢測電路 11’ 指紋資料採集模塊 12’ 應用處理器 121’ 資料處理單元 122’ 儲存單元 123’ 識別單元 2’ 光檢測器電路 21’ 光二極體感測單元 3’ 智能手機 31’ 觸控顯示屏幕<Acquaintances> 1'Biometric detection circuit 11’ Fingerprint data collection module 12’ Application processor 121’ Data Processing Unit 122’ storage unit 123’ Identification Unit 2'Light detector circuit 21' Photodiode sensing unit 3’ Smart phone 31’ Touch display screen
圖1為習知的一種生物特徵採集與識別裝置的方塊圖; 圖2為包含本發明之一種生物特徵資料加密裝置的一生物特徵採集與辨識裝置的方塊圖; 圖3為本發明之一種生物特徵資料加密方法的流程圖; 圖4為本發明之生物特徵資料加密裝置的一加密單元的方塊圖; 圖5為本發明之8階阿達瑪矩陣的一示圖; 圖6為本發明之8階阿達瑪矩陣的另一示圖;以及 圖7為本發明之生物特徵資料加密裝置的一解密單元的方塊圖。Figure 1 is a block diagram of a conventional biological feature collection and recognition device; FIG. 2 is a block diagram of a biometric collection and identification device including a biometric data encryption device of the present invention; Figure 3 is a flow chart of a method for encrypting biometric data according to the present invention; 4 is a block diagram of an encryption unit of the biometric data encryption device of the present invention; Figure 5 is a diagram of the 8-order Hadamard matrix of the present invention; Fig. 6 is another diagram of the 8-order Hadamard matrix of the present invention; and FIG. 7 is a block diagram of a decryption unit of the biometric data encryption device of the present invention.
1:生物特徵採集與辨識電路1: Biometrics acquisition and identification circuit
11:生物特徵資料採集模塊11: Biometric data collection module
12:應用處理器12: Application processor
121:資料處理單元121: Data Processing Unit
122:儲存單元122: storage unit
123:加密單元123: encryption unit
124:特徵比對單元124: Feature comparison unit
125:解密單元125: Decryption Unit
2:檢測器電路2: Detector circuit
21:光二極體感測單元21: Optical diode sensing unit
3:智能手機3: smart phone
31:觸控顯示屏幕31: Touch display screen
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW109114234A TWI737280B (en) | 2020-04-28 | 2020-04-28 | Biometric data encryption device and method and information processing device using the method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW109114234A TWI737280B (en) | 2020-04-28 | 2020-04-28 | Biometric data encryption device and method and information processing device using the method |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI737280B TWI737280B (en) | 2021-08-21 |
TW202141310A true TW202141310A (en) | 2021-11-01 |
Family
ID=78283282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW109114234A TWI737280B (en) | 2020-04-28 | 2020-04-28 | Biometric data encryption device and method and information processing device using the method |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI737280B (en) |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE69638018D1 (en) * | 1995-02-13 | 2009-10-15 | Intertrust Tech Corp | Systems and procedures for managing secure transactions and protecting electronic rights |
CA2836472C (en) * | 2011-05-18 | 2022-11-22 | Nextgenid, Inc. | Multi-biometric enrollment kiosk including biometric enrollment and verification, face recognition and fingerprint matching systems |
US10068118B2 (en) * | 2014-02-25 | 2018-09-04 | Tara Chand Singhal | Apparatus and method for a biometric sensor in a handheld mobile wireless communication device |
CN109842491B (en) * | 2017-11-28 | 2021-08-24 | 上海耕岩智能科技有限公司 | Electronic equipment |
CN109558799B (en) * | 2018-10-26 | 2022-04-15 | 国微集团(深圳)有限公司 | Fingerprint acquisition method and device and product adopting fingerprint acquisition device |
CN110276328B (en) * | 2019-06-27 | 2021-11-02 | Oppo广东移动通信有限公司 | Fingerprint identification method and related product |
CN110956468B (en) * | 2019-11-15 | 2023-05-23 | 西安电子科技大学 | Fingerprint payment system |
CN111027404B (en) * | 2019-11-15 | 2023-05-23 | 西安电子科技大学 | Fingerprint identification method based on fingerprint protection template |
-
2020
- 2020-04-28 TW TW109114234A patent/TWI737280B/en active
Also Published As
Publication number | Publication date |
---|---|
TWI737280B (en) | 2021-08-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20080005578A1 (en) | System and method for traceless biometric identification | |
US20100174914A1 (en) | System and method for traceless biometric identification with user selection | |
Ikuesan et al. | Digital behavioral-fingerprint for user attribution in digital forensics: Are we there yet? | |
KR20060018839A (en) | Smart authenticating card | |
Maiorana et al. | User adaptive fuzzy commitment for signature template protection and renewability | |
Belkhede et al. | Biometric mechanism for enhanced security of online transaction on Android system: A design approach | |
Prasad et al. | A study on multifactor authentication model using fingerprint hash code, password and OTP | |
Singh et al. | Cancelable iris template generation by aggregating patch level ordinal relations with its holistically extended performance and security analysis | |
Fang et al. | HandiText: Handwriting recognition based on dynamic characteristics with incremental LSTM | |
Lee et al. | Patenting trends in biometric technology of the Big Five patent offices | |
Rassomakhin et al. | Mathematical model for the probabilistic minutia distribution in biometric fingerprint images | |
Ara et al. | An efficient privacy-preserving user authentication scheme using image processing and blockchain technologies | |
Lalithamani et al. | Palm and hand vein-based fuzzy vault generation scheme for multibiometric cryptosystem | |
Malatji et al. | Acceptance of biometric authentication security technology on mobile devices | |
TWI737280B (en) | Biometric data encryption device and method and information processing device using the method | |
Alotaibi et al. | A novel transparent user authentication approach for mobile applications | |
Galbally et al. | Introduction to presentation attack detection in fingerprint biometrics | |
CN111709312B (en) | Local feature face recognition method based on combined main mode | |
Islam et al. | Technology review: image enhancement, feature extraction and template protection of a fingerprint authentication system | |
CN114444047A (en) | Identity authentication method, device, equipment and storage medium based on virtual reality | |
Terfa et al. | Multi-modal biometrics systems: Concepts, strengths, challenges and solutions | |
Talib et al. | Design and develop authentication in electronic payment systems based on IoT and biometric | |
SP et al. | Cancelable biometric scheme based on dynamic salting of random patches | |
Aithal | A Study on Multifactor Authentication Model Using Fingerprint Hash Code, Password and OTP | |
Dhiman et al. | An introduction to deep learning applications in biometric recognition |