201020938 九、發明說明: 【發明所屬之技術領域】 本發明涉及一種身份識別系統及方法,特別涉及一種 利用虹膜識別用戶身份的系統及方法。 【先前技術】 隨著社會的發展,身份識別的重要性正日益顯現,而 傳統的身份識別方式由於其固有的局限性已經遠遠不能滿 足要求’錄匙、卡片和身份證等容易丟失和偽造,密碼則201020938 IX. Description of the Invention: [Technical Field] The present invention relates to an identification system and method, and more particularly to a system and method for identifying a user identity using an iris. [Prior Art] With the development of society, the importance of identity recognition is becoming more and more obvious, and the traditional identification method is far from satisfying the requirements because of its inherent limitations. 'Log keys, cards and ID cards are easy to lose and forge. Password
容易遺忘’更為嚴重的是這些傳統的身份識別方式無法區 分真正的擁有者和取得身份標諸物的冒充者,一旦他人獲 得了這些身份標誌物,就可以享有與擁有者相同的權力。 近年來’基於生物特徵的識別技術應運而生,例如, 指紋識別方式。然而,由於指紋容易被破壞,因而也具有 不穩定性。 ' 【發明内容】 繁於以上内容,有必要提供—種虹膜朗系統及方 其能虹賴生物特性增強身份朗的精確性及穩 性。 裝置二裝:電子裝”, 識別碼的各用戶的===述的記憶體中儲存有作為 括:獲取模組,的特徵值,該系統包 並獲取相應的認證虹_^目機模組拍攝用戶的知臈, 認證虹_像崎處理;=組,用於對所獲取的 膜特徵提取模組,用於在處理 7 201020938 後的認證虹_像巾提取賴虹膜的特徵點,並計算所提 取的認證虹膜各舰‘_特徵值;及比對模组,用於 所提取的認證虹膜的魏點㈣徵值是否與減體中的識 別瑪相符’並在所提取的職虹膜的特徵點的特徵值與記 憶體中的朗碼相符時確賴用戶通過身份識別。 一種虹膜識财法,顧於電子裝置^,所述的電子 裝置包括相機模組及記顏,該方法包括如下步驟:將各 :戶的原始虹膜的特徵點的特徵值儲存至所述的記憶體 :虹=的相機模組拍攝用戶的虹膜,並獲取相應的認 故虹膜圖像;對所獲取的認證虹膜圖像進行處理;在處理 後的認證城圖料躲贿虹膜的舰點,並計算所提 取的認證蝴各賴_魏值;輯所提取的認證虹膜 =特徵點的特徵值是料記龍中的朗碼相符;及若所 奸取的…丑虹膜的特徵點的特徵值與記憶體中的識別碼相 付,則確認該用戶通過身份識別。 φ 相較於S知技術,所述的虹膜識別系統及方法可利用 t膜的生物特H (例如,虹膜的唯—性、穩定性等)增強 身份識別的精確性及穩定性。 【實施方式】 it發明膜識別系統較佳實施方式的運 2衣境及功纟b模組圖嘴述的虹膜識別系統2安裝並運行在 裝置1+。所述的電子裝置1可以是手機、PDA(personal digital assistant,個人鉍私虹、 人數位助理)、掌上遊戲機、數碼相機 或旱上電腦等可檇式電早 电千裝置,也可以是固定的門禁系統。 8 201020938 所述的電子裝置1還包括相機模組10、處理器以及記 憶體14。所述的相機模組10通常包括圖像感測器、鏡頭及 鏡座,也可包括其他元件或裝置。該圖像感測器可以是 CCD(Charge Coupled Device,電荷藕合器件),或者是 CMOS(C〇mplementary Metal-Oxide-SemiconductorIt is easy to forget. What is more serious is that these traditional methods of identification cannot distinguish between true owners and impostors who obtain identity objects. Once others have obtained these identity markers, they can enjoy the same rights as owners. In recent years, biometric-based recognition technology has emerged, for example, fingerprint recognition. However, since the fingerprint is easily destroyed, it is also unstable. [Explanation] In the above, it is necessary to provide an iris-lang system and its ability to enhance the accuracy and stability of the identity. Device 2: electronic device", each user of the identification code === stored in the memory stored as a feature value of the acquisition module, the system package and obtain the corresponding certification rainbow _ ^ eye machine module Shooting the user's knowledge, the certification rainbow _ _ _ processing; = group, for the acquired film feature extraction module, used to process the feature points of the iris _ iris after processing 7 201020938, and calculate The extracted irises of each of the certified iris '_ eigenvalues; and the comparison module are used to determine whether the Wei (4) eigenvalue of the extracted iris is consistent with the identification of the subtracted body and the characteristics of the extracted iris When the eigenvalue of the point coincides with the language in the memory, the user is identified by the identity. An iris identifiable method, the electronic device includes a camera module and a face, and the method includes the following steps : storing the feature values of the feature points of each household's original iris to the memory: the camera module of the rainbow=photographs the iris of the user, and acquires the corresponding iris image of the recognition; the iris image obtained for the authentication Like processing; after processing The card city map hides the ship's point of the iris, and calculates the extracted authentication butterfly _ wei value; the extracted identifiable iris = feature point eigenvalue is the lang code in the material record dragon; and if the traitor Taking the eigenvalue of the feature point of the ugly iris and the identification code in the memory, it is confirmed that the user is identified by the identity. φ Compared with the S-known technology, the iris recognition system and method can utilize the t-film The bio-specific H (for example, the uniqueness, stability, etc. of the iris) enhances the accuracy and stability of the identification. [Embodiment] It is a preferred embodiment of the invention to identify a membrane identification system. The iris recognition system 2 is installed and operated on the device 1+. The electronic device 1 may be a mobile phone, a PDA (personal digital assistant), a handheld game console, a digital camera or a dry camera. The electronic device 1 further includes a camera module 10, a processor and a memory 14. The camera module 10 is generally provided by a computer such as a portable electric early thousand device or a fixed access control system. Including image sensing Lens, and the lens holder, may also include other elements or devices. The image sensor may be a CCD (Charge Coupled Device, Charge Coupled Device) or CMOS (Metal-Oxide-Semiconductor C〇mplementary
Transistor,互補型金屬氧化物半導體)。在本較佳實施方式 中,所述的相機模組10用於拍攝用戶的虹膜進行識別。所 述的處理器12用於執行所述虹膜識別系統2以及所述電子 ❹ 裝置1内的其他應用程式’例如:作業系統等。 所述的記憶體14可以内置於所述電子裝置卫本身的記 憶體,還可以是外部儲存卡,如SM卡(Smart Media Card, 智慧媒體卡)、SD卡(Secure Digital Card,安全數位卡) 等。所述的記憶體14用於儲存各類資料,包括利用所述相 機模組10拍攝的虹膜的圖像、利用所述的虹膜識別系統2 對虹膜的圖像處理後所得到的資料等。 ❹ 所述的虹膜識別系統2主要包括獲取模組20、處理模組 22、特徵提取模組24、儲存模組26以及比對模組28 ^ 一在通過識別用戶虹膜以驗證用戶身份之前,需要先建 立一個儲存有各用戶的原始虹膜圖像及相關 資料的虹膜特 資料庫,以便在後續的識別過程中進行比對從而驗證用 戶身份。 所述的獲取模組2〇與所述的相機模組1〇相連用於令 Z述的相機模_拍攝用戶的虹膜,並獲取相應的原始虹 201020938 所述的處理模組22,用於對所獲取的原始虹膜圖像進 行處理’包括:將原始虹_像由彩色變成黑白色;降低 原始虹膜圖像的亮度;增加縣虹膜®像的對比度。Transistor, complementary metal oxide semiconductor). In the preferred embodiment, the camera module 10 is used to capture the iris of the user for recognition. The processor 12 is configured to execute the iris recognition system 2 and other applications within the electronic device 1 such as an operating system or the like. The memory 14 can be built in the memory of the electronic device itself, or can be an external storage card, such as a SIM card (Smart Media Card), an SD card (Secure Digital Card). Wait. The memory 14 is used for storing various types of data, including an image of an iris photographed by the camera module 10, data obtained by processing the image of the iris by the iris recognition system 2, and the like. The iris recognition system 2 mainly includes an acquisition module 20, a processing module 22, a feature extraction module 24, a storage module 26, and a comparison module 28, which are required to identify the user's iris to verify the identity of the user. First, an iris special database storing the original iris images and related materials of each user is established to perform comparison in the subsequent identification process to verify the identity of the user. The acquiring module 2〇 is connected to the camera module 1〇 for the camera module of the Z to capture the iris of the user, and obtain the corresponding processing module 22 described in the original rainbow 201020938, for The acquired original iris image is processed 'including: changing the original rainbow image from color to black and white; reducing the brightness of the original iris image; increasing the contrast of the county iris image.
所述的特徵提取模組24,用於在處理後的原始虹膜圖 像中提取縣虹_特徵點,並計算所提取的原始虹膜各 特徵點的特徵值。其中’所述的原始虹膜的特徵點包括主 特徵點與次特徵點,所述的特徵點的特徵值包括,但不限 於,特徵點數4、雜點絲、特徵料雜及特徵點距 π闽句虹膜結稱的簡單示意圖,其中,虹 位於眼睛黑色瞳孔7和白色鞏膜(圖中、> 狀部分,總體上呈現一種由,丨休 μ s的圓〕 、“ 禋由裏到外的放射狀結構,由相, m織構成’包含有报多相互交錯的類似… 點、細絲、冠狀、條紋、隱窝等細節特徵 方式中,以隱窩為特徵點,故圖3中 實> 明本較娜m編 紋等特徵未作繪製。 ,、田…冠狀、1 在本較佳實施方式巾,所述的原始虹膜 特徵點與捕徵點,其巾,縣縣括, 离作為主特徵點,如圖4、圖5中所:中面積最大的-個丨 點,而;f 0虹#中a不的方框部分為主特1 而原始虹膜^其餘的隱离作為次特徵點, 中所示的圓圈部分為次特徵點。在圖4 圖 原始虹旗的主特徵點的數量為1,次特徵點的數不量=中 所述的_取模組24_如膜各特徵點的座; 201020938 示。提取原始虹膜的主特徵點和次特徵點 座標圖可以計算^點上建立座標圖,然後根據該 t算出原始虹膜各特徵點的座標。 如圖5所相雜跡触24料_域各賴點的形狀 不,將每個特徵點與其距離最近的特徵點相連: 計算出各特徵點之間的=所計算出的各特徵點的座襟, 原始====所=將特徵提取模組24所提取的 "及所叶算的各特徵點的特徵值儲存至 虹臈特徵資料庫,作為電子裝置_:戶 別碼以識別-個或者多個用戶的身份。 飞者多組識 所述2='用戶身份時’所述的獲取模組20還用於令 所述的相機模組_攝用戶的虹膜(為便於描述,; ❿ :=r的虹膜統稱為“認證虹膜,,),並獲取相應 所述的處理模組22,還用你料祕權仏t 進行處理’包括:_證虹_二=:圖: 低認===;増加認證虹膜圖像的對比度。 所述的特徵k取模組24,用於在處 像中提取認證虹_特徵點,並計算所提 特徵點的特徵值。其中,所述 犋各 方法與原始虹膜特徵點的心相同 11 201020938 的特徵點的特徵值的計算方法與原始虹膜特徵點的 的計算方法相同。 、、徵值 所述的比對模組28,用於比對所述的認證知膜特徵點 的特徵值是否與記憶體14中的虹膜特徵資料庫的識別瑪相 符。若所述的虹膜特徵資料庫包括多組識別碼,則判斷= 述的認證虹膜特徵點的特徵值是否與其中一組識別碼相 符。 Ο 其中,比對標準可以根據特徵值的各參數 例如,比對標準包括,但不限於如下細項:細項^㈣ 證虹膜的主特徵點與作為識別瑪的其中一個原始虹膜= 特徵點的座標_,料確認所述認證虹膜的主特徵 該原始虹膜的主特徵點相同;細項二,當認證虹膜的次特 徵點的數量達龍原始虹膜的次特徵點數量的以 認,與該原始虹膜在次特徵點的數量上相 膜的各‘徵==::=的座標與該原始虹 膜與該朴_ ’财認為認證虹 的細項可根據實降要书 τ棵準 貫際要求進仃設置、新增、減少或者修改。 點的特徵值進還用於當所述的認證虹膜的特徵 特徵點的特徵值與識二=二,確認該認證虹臈的 別。 付並確s忍該用戶通過身份識 如圖2所示,是本發明 程圖。首先,步驟S2,所心識财法較佳實施方式的流 *·的獲取模組20令所述的相機模 12 201020938 組ι〇拍攝多個用$ μ丄 用戶的虹膜,並獲取相應的原始虹膜圖像。 + ’所述的處理雜22對所獲取的原始虹 進仃處理,包衽.脸e u i 豕 .匕栝.將原始虹膜圖像由彩色變成黑白色;降 膜圖像的亮度;增加原始虹膜圖像的對比度。 liut ,所述的特徵提取模組24在處理後的原始虹膜 々提取原始虹膜的特徵點,並計算所提取的原始虹膜 寺徵點的特徵值。其中,所述的原始虹膜的特徵點包括 主特徵點與次特徵點,所述的特徵點的特徵值包括,但不 限於特徵點數量、特徵點座標、特徵點的形狀及特徵點 距離。 ” 在本較佳實施方式中,所述的原始虹膜特徵點包括主 特徵點與次特徵點,其中,原始虹膜中面積最大的一個隱 窩作為主特徵點,如圖4、圖5中所示的方框部分為主特徵 點,而原始虹膜中其餘的隱窩作為次特徵點,如圖4、圖5 中所示的囷圈部分為次特徵點。 步驟S8,所述的儲存模組26將所提取的原始虹膜各特 徵點以及所計算的各特徵點的特徵值儲存至記憶體14内, 作為電子装置1識別各用戶身份的識別碼》 步驟Sl〇,所述的獲取模組20令所述的相機模組1〇拍 攝待識別身份的用戶的虹膜’並獲取相應的認證虹膜圖像。 步驟S12,所述的處理模組22對所獲取的認證虹膜圖 像進行處理,包栝:將認證虹膜圖像由彩色變成黑白色; 降低認證知膜圖像的亮度;增加認證虹膜圖像的對比度。 步驟S14,所述的特徵提取模組24在處理後的認證虹 13 201020938 膜圖像中提取認證虹膜的特徵點,並計算所提取的認證虹 膜各特徵關特徵值。其巾,所述的認證虹膜的特徵點的 提取方法與原始虹膜特徵點的提取方法㈣,所述的認證 虹膜的特徵點的特徵㈣計算方法與原始虹麟徵點的特 徵值的計算方法相同。The feature extraction module 24 is configured to extract a county rainbow feature point in the processed original iris image, and calculate feature values of the extracted original iris feature points. The feature points of the original iris include a primary feature point and a secondary feature point, and the feature values of the feature point include, but are not limited to, feature point number 4, noise point, feature material miscellaneous, and feature point distance π A simple schematic diagram of the haiku iris, in which the rainbow is located in the black pupil 7 and the white sclera (in the figure, the > part, generally presents a circle of 丨 μ μ s), and 禋 from inside to outside Radial structure, composed of phase, m-woven, contains similar features such as points, filaments, crowns, stripes, crypts, etc., with crypts as feature points, so Figure 3 is real > The characteristics of Mingben and Nam are not drawn. ,, Tian... crown, 1 in the preferred embodiment of the towel, the original iris feature points and the capture points, the towel, the county, including The main feature points are as shown in Fig. 4 and Fig. 5: the largest area in the middle area, and the frame part of a not in f 0 rainbow# is the main feature 1 and the original iris is the sub feature. The circle part shown in the point is the sub-feature point. In Figure 4, the main feature point of the original rainbow flag The number of sub-features is 1, the number of sub-feature points = the module _ taken in the module 24_ such as the seat of each feature point of the film; 201020938 shows that the main feature points and sub-feature points of the original iris can be extracted to calculate ^ A coordinate map is created on the point, and then the coordinates of the feature points of the original iris are calculated according to the t. As shown in Fig. 5, the shape of each of the points is not connected, and each feature point is connected to the feature point closest to the distance. : Calculate the coordinates of the calculated feature points between the feature points, the original ===== the feature value of each feature point extracted by the feature extraction module 24 and the leaf The storage module is stored in the rainbow trout feature database as an electronic device_: household code to identify the identity of one or more users. The acquisition module 20 is also used by the plurality of participants to identify the 2='user identity' In the camera module described above, the user's iris is photographed (for convenience of description; the iris of the ❿:=r is collectively referred to as "authentic iris,"), and the corresponding processing module 22 is obtained, and the secret is also used by you. The weight 仏t is processed 'including: _ 证虹_二=: Figure: Low recognition ===; 増 plus the contrast of the certified iris image. The feature k is taken by the module 24 for extracting the authentication rainbow feature point in the image, and calculating the feature value of the proposed feature point, wherein the method is the same as the heart of the original iris feature point 11 201020938 The calculation method of the feature value of the feature point is the same as the calculation method of the original iris feature point. The comparison module 28 for locating the value is used to compare whether the feature value of the authentication film feature point and the memory are compared. The identification of the iris feature database in the body 14 is matched. If the iris feature database includes a plurality of sets of identification codes, it is determined whether the feature value of the certified iris feature point corresponds to one of the identification codes. The comparison criteria may be based on various parameters of the eigenvalues, for example, the comparison criteria include, but are not limited to, the following items: the fine item ^(4) the main feature point of the iris and one of the original irises as the identification mas = the coordinates of the feature point _ The main feature of the original iris is confirmed to be the same as the main feature of the original iris; the second item, when the number of sub-feature points of the certified iris is up to the number of sub-feature points of the original iris, With the original iris in the number of sub-feature points, the coordinates of the respective signs of the sign ==::= and the original iris and the pak _ 'cai think that the details of the certified rainbow can be based on the actual drop of the book τ It is required to set up, add, reduce or modify. The feature value of the point is further used for confirming the identity of the rainbow trout when the feature value of the characteristic feature point of the authentication iris is the same as the identification value. Pay and confirm that the user is identified by the identity shown in Figure 2, which is the process diagram of the present invention. First, in step S2, the acquisition module 20 of the preferred embodiment of the preferred embodiment of the method of the financial method causes the camera module 12 201020938 group to photograph multiple irises of the user with $ μ丄 and obtain the corresponding original Iris image. + 'The processing of the miscellaneous 22 pairs of the obtained original rainbow 仃 processing, including 脸. face eui 豕. 匕栝. The original iris image from color to black and white; the brightness of the falling film image; increase the original iris map Like the contrast. Liut, the feature extraction module 24 extracts the feature points of the original iris in the processed original iris, and calculates the feature values of the extracted original iris temple points. The feature points of the original iris include a primary feature point and a secondary feature point, and the feature values of the feature point include, but are not limited to, the number of feature points, the feature point coordinates, the shape of the feature points, and the feature point distance. In the preferred embodiment, the original iris feature points include a primary feature point and a secondary feature point, wherein a crypt having the largest area in the original iris is used as the main feature point, as shown in FIG. 4 and FIG. 5 . The box portion is the main feature point, and the remaining crypts in the original iris are used as the sub-feature points, and the loop portion shown in Fig. 4 and Fig. 5 is the sub-feature point. Step S8, the storage module 26 The extracted feature points of the original iris and the calculated feature values of the feature points are stored in the memory 14 as an identification code for identifying the identity of each user by the electronic device 1 Step S1, the obtaining module 20 The camera module 1 captures the iris of the user to be identified and acquires the corresponding authentication iris image. Step S12, the processing module 22 processes the acquired authentication iris image, including: The certified iris image is changed from color to black and white; the brightness of the certified film image is lowered; the contrast of the certified iris image is increased. Step S14, the feature extraction module 24 is processed after the authentication rainbow 13 201020938 film image in Taking the feature points of the certified iris, and calculating the extracted feature values of the extracted irises, the method for extracting the feature points of the certified iris and the method for extracting the original iris feature points (4), the authentication iris The feature of the feature point (4) is calculated in the same way as the eigenvalue of the original Honglin sign.
步驟S16,所述的比對模組28比對所述的認證虹膜特 徵點的特録是錢記,_14巾社難徵資料庫的一組 識別碼相符。其巾,比對鮮可以根據魏值的各參數進 行設置。若所述的認證紅膜特徵點的特徵值與記憶體咐 的識別碼相符,則執行步驟S18,否則,結束本流程。 步驟S18,所述的比對模組28確認該用戶通過身份識 別,並結束流程。 綜上所述’本發明符合發明專利要件,轰依法提出專 利申請。惟,以上所述者僅為本發明之較佳實施方式,本 發明之範圍並不以上述實施方式為限,舉凡熟悉本’案技藝 之人士援依本發明之精神所作之等效修飾錢化,皆應 蓋於以下申請專利範圍内。 ‘/ 【圖式簡單說明】 圖1是本發明虹膜識別系統的較佳實施方式的運用产 境及功能模組圖。 & 圖2是本發明虹膜識別方法較佳實施方式的流程圖。 圖3是本發明虹膜識別系統的虹膜結構示意圖。 圖4是本發明虹膜識別系統的虹膜特徵點座標圖。 圖5是本發明虹膜識別系統的虹膜特徵點間的相對位 201020938 置的示意圖。 【主要元件符號說明】 電子裝置 1 相機模組 10 處理器 12 記憶體 14 虹膜識別系統 2 獲取模組 20 處理模組 22 特徵提取模組 24 儲存模組 26 比對模組 28 虹膜 5 隱窩 6 瞳孔 7 15In step S16, the comparison module 28 compares the special signature of the authentication iris feature point with a set of identification codes of the _14 towel society database. For the towel, the comparison can be set according to the parameters of the Wei value. If the feature value of the authentication red film feature point matches the identification code of the memory port, step S18 is performed; otherwise, the flow is ended. In step S18, the comparison module 28 confirms that the user recognizes by identity and ends the process. In summary, the invention conforms to the patent requirements of the invention, and the patent application is filed according to law. However, the above description is only a preferred embodiment of the present invention, and the scope of the present invention is not limited to the above-described embodiments, and those skilled in the art will be able to use the equivalent modification of the spirit of the present invention. All should be covered by the following patent application. ‘/ [Simple Description of the Drawings] Fig. 1 is a diagram showing the operational environment and functional modules of the preferred embodiment of the iris recognition system of the present invention. & Figure 2 is a flow chart of a preferred embodiment of the iris recognition method of the present invention. 3 is a schematic view showing the structure of an iris of the iris recognition system of the present invention. 4 is a graph showing the coordinates of an iris feature point of the iris recognition system of the present invention. Figure 5 is a schematic illustration of the relative position of the iris feature points of the iris recognition system of the present invention 201020938. [Main component symbol description] Electronic device 1 Camera module 10 Processor 12 Memory 14 Iris recognition system 2 Acquisition module 20 Processing module 22 Feature extraction module 24 Storage module 26 Comparison module 28 Iris 5 Crypt 6 Pupil 7 15