1334578 9έ·1〇_31第96⑶8〇8號專利說明書、面式及申靖專利範圍修正本 九、發明說明： 【發明所屬之技術領域】 本發明係關於-種指紋及掌紋雙特徵辨識方法 是關於利用操取之手掌紋及手指紋之影像設徵巴 域’再加以比對，以提升執行速度及辨識 = 辨識方法及其裝置。 王物特徵 【先前技術】 個人辨識系統近年來快速的發展，因此生物獨 -特徵被做為辨識系統的f要辨識特徵，不管是針對於 瞳孔、聲音等都有許多人在作研究，然而對於以 的特徵如何作最麵確認也非常重要，但是現有的方 P有-疋的複雜度因而拉長其確認所需的時間。、 —穿習用辨識方法，如中華民國公告第476917 :旱生物特徵確認系統」發明專利，其包含— =且、-影像分割模組、-特徵揭取模組及至少：特』 =組。其彻該影賴取歡將—手掌影像進 分割模組標示出手掌掌心與指節間的ΐ 同解析产的!特徵練模組抽取出數個掌形特徵及數個不 的比對;=^ ;最後再藉由舰確認模組進行特徵 者的身分。以進一步輸出辨識結果，進而可確認使用 驟中，I將用具有不少缺點’例如在特徵擁取的步 、咖〕：特定線段的灰階資料不僅透過小波〔 、以求得數個不同解析度的特徵向量，接著 1334578 96·10·31第96131808號專利說明書、圓式及申請專利範圍修正本 再I呈由主轴分析iPCA〕的轉換以降低該特徵向量的維产 /而，其將造成影像轉換複雜度的提升， 二 整體辨識及tb制咖。 州於降低 有鑑於此，本發明改良上述之缺點，其 手指紋影像沿設定的基準線進行灰階化處理;'並判 灰階的變化在手掌紋及手指紋之位置上設定數個 ;將該特徵區域與-主機⑽紋路特徵資料庫藉由；識 方法進行比對後，再將比對結果進行輸出，因此能提升執 行速度及辨識效果。 开執 【發明内容】 本發明之主要目㈣提供—翻紋及掌紋雙特徵 識方法，其係在所掏取之手掌紋及手指紋影像設定數個特 徵區域，藉由影像辨識與已建立於一紋路特徵資料庫進行 比對，使得本發明具有提升執行速度及觸效果之功效。 根據本發明之指紋及掌纹雙特徵辨識方法，取— 待=手掌紋及手指紋之影像，並沿選取之基準線進行灰階 化處理’崎料掌紋及手驗之影絲生至少—線段灰 階分佈圖，依據該線段灰階分佈圖設定至少一特徵區域， ^該特徵區域經由-辨識方法與一纹路特徵資料庫進行 【實施方式】 為發明之上述及其他目的、特徵及優點能更明顯 ’下文特舉本發明之較佳實施例，並配 作詳細說明如下·· 圓式 —6 — 1334578 • ‘ ‘ο.31第__號專利說明書 '明式及申請專利範团修正本 eU1至3圖所7’其揭示本發明應用於較佳實 • 施狀指缺掌时贿_綠^置，料包含一二 . 房卜一影像摘取單元2及-主機3。該暗房！之外周係 ，以便在其内部界定形成—可隔絕外部光 .、線之工間〔未標示〕，且該暗房1内部係設有一定 ^該影_取單元2錢置在該暗房i内部，而較佳^ 5又置在相對於該定位平台1〇的 ’、 手掌及手射打放置錢^"： ^相者之待測 〇〇 , 〇 _ 疋位十口 10之上，而使該影像 掘取早70 2擷取職測手掌紋及手·之影像。 請再參照第1至3圖所示，本發明應用於較佳實施例 之指紋及掌紋雙特徵辨識方法之影像摘取單元2係包含一 二光源21及_移動裝置22。該攝像機2〇係對 應权置在該疋位平台1〇之下方，以便攝錄該定位平二1〇 上掌紋或手指紋之局部影像，該攝像機心口選自 感光耦。兀件型〔Charge c〇u_，c f充性：化金屬半導體型〔c〇mplementary m时⑽e Semiconductor，CMOS〕之數相潘旦《嬰从斗 蔽外帛六社― 數位攝〜态材。該光源21係對 位平台1〇之下方’且可產生-光線，以照射 ===該移動最置22係樞接該攝像機2。，該 2 = 馬達，或一旋轉馬達與齒輪組之 二 袭置22帶動該攝像機2〇進行平移 之作動，進而使該攝像機20可藉由移動而 紋及手指紋之全部影像。 ^付利于旱 請再參照第1至3圖所示，本發明應用於較佳實施例 31第96131808號專利說明書、圓式及申請專利範圍修正本 之指紋及掌紋雙特徵辨識方法之主機3較佳係可選自一個 亡電腦’該主機3透過-影賴取卡31而連接該影像拍員取 2之攝像機2〇，以便該主機3透過該攝像機2〇進行 曰紋及手彡像的存取、觸及輯作業·並且該主 機亦連接該移動裝置22，使該主機3可控制該移動裝置22 之平，位移量，進而控制該攝像機2()之移動位置。該主機 内部係具有一紋路特徵資料庫，以便該待測手掌紋及手 指紋進行比對作業。 再者’該指紋及掌紋雙特徵辨識系統在該 〔未繪示〕，其可攝錄該待測者之人像，且 i一夕/又置之攝像機係—連接該主機3，以供該待測者於 路2測試時’可將人像及待測者之個人資訊儲存於該紋 路特徵貢料庫内。 雙特第4圖所示’本發明較佳實施例之指紋及掌紋 2^識方法係包含步驟：〔S1〕利用該影像操取單元 測手掌紋及手指紋之影像，並將該影像藉由該 ^卡31輸入至該主機3;〔S2〕在該主機3中針對 二測手掌紋及手指紋之影像選取至少—基準線U、丄2 階化^基準線L1、U進行至少—手指紋及手掌紋之灰 綠，並產生至少—線段灰階分佈圖；⑶〕藉由該 徵階分佈圖分別針對手指紋及手掌紋位置設定數個特 的辨識，’ ^S4〕將該特徵區域經由一辨識方法進行特徵 步狀勃/、該紋路特徵資料庫進行比對。藉此，由上述 行可快速且精確的判斷並辨識該使用者之身分。 96.10.31第96131808號專利說明書'明式及申請專利範圍修正本 纹雔2參㈣1 _示，本發明較佳實施例之指紋及掌 之ΐ像〔S1〕係將待測手掌紋及手指紋 像輸入該主機3進行儲存及：二卡31 _ ’再將影 及士麻誓 存便與資料庫進行後續的辨識 像n、’且#機3可藉由—人機介面將獅後的影 ^不，以便操作人員可更加人性化的使用操作環境 ’進而能找麵掌握f彡像的練資訊。 ^照第4至9 _示，本發明較佳實施例之指紋及 =雙特ί辨識方法之步驟〔S2〕係藉由該主機3預先在 =，測手掌紋及手指紋之影像選取數條基準、線U、L2,在 T手才曰、，文位置之基準線u係連接手指影像上邊緣之中 :點及下邊緣之中心點所界^ ;另外’將中指影像一半的 i方设為中點’繪製通過該中點且垂直的線段，以界定形 成在手掌位置之基準線L2。 °月再參照第4至9圖所示，本發明較佳實施例之指紋 及掌紋雙特徵辨識方法之步驟〔S2〕之灰階化處理主要係 針=手掌紋及手指紋位置選取的基準線L]i、l2利用線段 =1%直方圖的方式進行灰階化分佈作業，以便透過該人機 1^刀別獲得對應的線段灰階分佈圖，藉以觀察其線段灰 階分佈情形。例如第5及7圖分別的下端部所示的線段灰 階^佈情形及中線灰階分佈情形係分別針對單一手指紋及 手掌纹之基準線U、L2所分析出的線段灰階分佈圖。. a請再參照第4至9圖所示’本發難佳實施例之指紋 及掌故雙特徵辨識方法之步驟〔S3〕係將上述線段灰階分 —9 — 1334578 9έ·10·31第96131808號專利說明書、圖式及申請專利範圍修正本 - 佈圖進一步作分析。主要將線段灰階分佈圖中，數個特定 ' 區間的辨識區域尺1至R4將發生灰階分佈的變化作為對應 . 特徵區域的選取，該灰階分佈的變化之依據係指灰階值低 的部分。更詳言之，例如，第5圖的下端部所顯示的線段 灰階分佈情形在160至175、325至340及510至525三個 . 辨識區域幻至R3的灰階值與周圍之灰階值有明顯的差異 ，藉由找出該三個辨識區域R1至尺3中個別最低的灰階分 φ 佈位置，亦即在160、326及515之位置作為單一手指的指 紋之3個對應特徵區域之選取；又，例如第7圖的下端部 所顯示的中線灰階分佈情形在25〇至275具有該辨識區域 R4，同樣在該辨識區域R4内選取最低的灰階分佈位置作 為手掌的指紋之對應特徵區域之選取，事實上，該特徵區 域之選取係最接近掌心的三條主線中的中間主線之位置。 ‘其中該灰階分佈的變化主要係因為單一手指中都具有3個 較明顯的皺摺〔即較深遂的皺摺〕，因此與該手指其他部 • 位將具有灰階的變化，各手指具有3個特徵區域，因此， 在此步驟中’於手指指紋的部分可選取至少15個特徵區域 ，再以該特徵區域之中心點為準，針對原始的各個手指紋 影像設定具有60x60像素點〔pixel〕之特徵區域“丨至“] ' 。同樣的，如第8圖所示，針對手掌紋的特徵區域之中心 - 點為準向左右各設定125個像'素點、向上設定75個像素點 及向下設定175個像素點，以形成一具有25〇χ25〇像素點 的特徵區域M4，進而作為手掌紋的辨識範圍，為了獲得 更精確的辨識效果，可以該特徵區域M4為中心點，^在 —10 — 1334578 • 96.10.31第96131808號專利說明書、圚式及申請專利範圍修正本 其周圍另選取4個特徵區域M5 ’且該4個特徵區域 : 分別具有60x60像素點。 • 請再參照第1及4至9圖所示，本發明較佳實施例之 指紋及掌紋雙特徵辨識方法之步驟〔S4〕主要將上述所選 取之數鋪徵區域Ml至M5透過酬識方法，例如本發 明較佳實施例之相關係數法〔c咖lati〇n c〇efficiem〕逐 -與該主機3之紋路特徵資料勒的數個樣板之影像進行 φ 灰階相似程度之比對，以便各個特徵區域Ml至M5可對 紐得-輯分數’並與該域3所設定之—預設分數進 订比較’若該比對分數高於該預設分數，則比對結果為該 特徵區域Ml至M5之影像係符合對應樣板之影像。因此 ，本發明之指紋及掌紋雙特徵辨識方法可同時藉由手掌紋 及手指紋之選取的數個特徵區域M1至M5進行辨識，並 藉由該辨識方法進行比對，以進一步提升執行速度及辨識 效果。 • -如上所述，相較於習用技術具有整體辨識時間長及比 對時間長等缺點，本發明藉由該影像掏取單元2將手掌紋 及手指紋影像沿設定的基準線u、L2進行灰階化處理； ' 並藉㈣域3騎灰_變化在手掌紋及手指紋之位置 上選取數個特徵區域Ml至M5 ;將該特徵區域Ml至M5 • ㈣像触主機3内建立的樣板雜it行比紐，再將比 對結果進行輸出，其確實可提升執行速度及辨識效果。 、雖然本發明已利用上述較佳實施例揭示，然其並非用 以限定本發明，任何熟習此技藝者在不脫離本發明之精神 —11 — 1334578 96.10.31第96131808號專利說明書、圖式及申請專利範圍修正本 和範圍之内，相對上述實施例進行各種更動與修改仍屬本 . 發明所保護之技術範疇，因此本發明之保護範圍當視後附 之申請專利範圍所界定者為準。1334578 9έ·1〇_31 No. 96(3)8〇8 Patent Specification, Surface Type and Shenjing Patent Range Revision IX. Invention Description: The present invention relates to a fingerprint and palmprint dual feature identification method. About using the image of the palm of the hand and the image of the hand fingerprint to set up the map, and then compare it to improve the execution speed and identification = identification method and its device. Wangwu characteristics [Prior technology] The personal identification system has developed rapidly in recent years. Therefore, the biological unique feature is used as the identification feature of the identification system. Whether it is for pupil, sound, etc., many people are studying it. It is also important to know how the features are most identifiable, but the existing square P has the complexity of - and thus lengthens the time required for its validation. - wearing the identification method, such as the Republic of China Announcement No. 476917: Drought Biometrics Confirmation System" invention patent, which includes - = and - image segmentation module, - feature extraction module and at least: special" = group. It is the same as the analysis of the palm of the palm and the knuckle. The feature training module extracts several palm-shaped features and several pairs of comparisons; ^; Finally, the identity of the feature is carried out by the ship confirmation module. In order to further output the identification result, it can be confirmed that I will use a number of disadvantages, for example, in the step of the feature acquisition, the grayscale data of the specific line segment not only passes through the wavelet [, in order to obtain several different resolutions. The feature vector of degree, followed by the patent specification of 1334578 96·10·31 No. 96131808, the round and the patent scope revision, and the conversion of the spindle analysis iPCA] to reduce the yield of the feature vector, which will cause Image conversion complexity is improved, two overall identification and tb coffee. In view of the above, the present invention improves the above disadvantages, and the fingerprint image of the hand is gray-scaled along the set reference line; and the change of the gray scale is set in the position of the palm print and the hand fingerprint; The feature area and the host (10) texture feature database are compared by the identification method, and then the comparison result is output, thereby improving the execution speed and the recognition effect. The main object of the present invention is to provide a dual feature method for embossing and palmprint, which is to set a plurality of feature regions in the palmprint and hand fingerprint images captured by image recognition and has been established by A texture feature database is compared, so that the invention has the effect of improving the execution speed and the touch effect. According to the fingerprint and palmprint dual feature identification method of the present invention, the image of the palmprint and the hand fingerprint is taken, and the grayscale processing is performed along the selected reference line. 'The texture of the palm and the handprint are at least the line segment. a gray scale distribution map, wherein at least one feature region is set according to the gray scale distribution map of the line segment, and the feature region is performed by the identification method and a texture feature database. [Embodiment] The above and other objects, features and advantages of the invention can be further improved. Obviously, the preferred embodiment of the present invention is described below and is described in detail as follows. · Round - 6 - 1334578 • ' '. § _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ To the figure 7', it is disclosed that the present invention is applied to a preferred embodiment of the present invention, and the bribe_green is placed in the palm of the hand. The material includes one or two. The image is extracted from the unit 2 and the host 3. The darkroom! Outside the system, in order to define the formation inside - the external light can be isolated, the line of the work room [not marked], and the inside of the darkroom 1 is provided with a certain ^ the shadow _ take the unit 2 money placed inside the darkroom i, And the better ^ 5 is placed in the ', palm and hand shots relative to the positioning platform 1 ^ put the money ^ ": ^ the person to be tested, 〇 _ 疋 10 above 10, and so The image was taken as an early 70-inch image of the hand palmprint and the hand. Referring to Figures 1 to 3, the image pickup unit 2 of the present invention applied to the fingerprint and palmprint dual feature recognition method of the preferred embodiment comprises a light source 21 and a mobile device 22. The camera 2 is correspondingly positioned below the clamping platform 1 , to record a partial image of the palm print or the hand fingerprint of the positioning flat. The camera core is selected from the photosensitive coupling.兀 型 Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char Char The light source 21 is positioned below the alignment platform 1' and can generate - light to illuminate === the most mobile 22 is pivotally attached to the camera 2. The 2 = motor, or a rotating motor and the gear set of the second set 22 drives the camera 2 to perform a translational operation, thereby enabling the camera 20 to move and print the entire image of the fingerprint. Please refer to FIGS. 1 to 3 for further explanation. The present invention is applied to the host computer 3 of the fingerprint and palmprint dual feature identification method of the preferred embodiment 31, No. 96,131,808, the round and the patent scope modification. The system can be selected from a dead computer. The host computer 3 connects to the camera 2 to capture the camera 2 through the camera 31, so that the host 3 can store the crepe and the handcuff image through the camera 2 The host device is also connected to the mobile device 22, so that the host device 3 can control the level of displacement of the mobile device 22, thereby controlling the moving position of the camera 2 (). The internal body of the host has a texture feature database for the palm print and the hand fingerprint to be compared. Furthermore, the fingerprint and palmprint dual feature recognition system is in the [not shown], which can record the portrait of the test subject, and the i-night/reset camera system is connected to the host 3 for the waiting When the tester is on the road 2 test, the personal information of the portrait and the test subject can be stored in the texture feature tribute library. The fingerprint and palmprint 2 method of the preferred embodiment of the present invention includes the steps of: [S1] using the image manipulation unit to measure the image of the palm print and the hand fingerprint, and the image is used by The card 31 is input to the host 3; [S2] selects at least the reference line U, the 阶2 stepped reference line L1, U for at least the hand fingerprint in the host 3 for the image of the palm measuring palm and the hand fingerprint. And the gray color of the palm print, and at least - the gray scale distribution map of the line segment; (3)] by the scale map, a plurality of special identifications are set for the hand fingerprint and the palm print position respectively, and the feature region is passed through the ' ^S4 An identification method performs feature alignment on the feature step and/or the texture profile database. Thereby, the user can quickly and accurately determine and recognize the identity of the user. 96.10.31 Patent No. 96131808 'Ming and Patent Application Revisions 本 2 ( (4) 1 _ shows that the fingerprint and the palm image [S1] of the preferred embodiment of the present invention are to be tested palm prints and hand fingerprints For example, input the host 3 for storage and: 2 cards 31 _ 'and then the shadows and the swearing and the database for subsequent identification like n, 'and #机3 can be used by the human-machine interface to the shadow of the lion ^No, so that the operator can use the operating environment more humanely, and then can find the information of the f-image. According to the fourth to the ninth, the step (S2) of the fingerprint and the double-tele identification method of the preferred embodiment of the present invention is selected by the host 3 in advance, the image of the palmprint and the hand fingerprint is selected. The reference line, line U, L2, in the T hand, the reference line u of the text position is connected to the upper edge of the finger image: the center point of the point and the lower edge is bounded; and the other is the half of the middle finger image. A line segment passing through the midpoint and perpendicular is drawn for the midpoint to define a reference line L2 formed at the palm position. Referring to FIGS. 4 to 9 again, the grayscale processing of the step [S2] of the fingerprint and palmprint dual feature identification method according to the preferred embodiment of the present invention is mainly the needle=the palm line and the fingerprint fingerprint position selection reference line. L]i, l2 use the line segment=1% histogram to perform the gray-scale distribution operation, so as to obtain the gray-scale distribution map of the corresponding line segment through the human-machine, so as to observe the gray-scale distribution of the line segment. For example, the gray scales of the line segments shown in the lower end of the fifth and seventh graphs and the gray scale distribution of the midline are respectively the gray scale distribution of the line segments analyzed for the reference lines U and L2 of the single hand fingerprint and the palm print. . a Please refer to the steps of the fingerprint and the double feature identification method of the present invention as shown in Figures 4 to 9 [S3]. The gray line of the above line segment is divided into 9- 1334578 9έ·10·31 No. 96131808 The patent specification, schema and patent application scope revision - layout are further analyzed. In the gray-scale distribution diagram of the line segment, the change of the gray-scale distribution occurs in the identification area of the specific 'intervals> from 1 to R4. The selection of the characteristic area is based on the change of the gray-scale value. part. More specifically, for example, the grayscale distribution of the line segments displayed at the lower end of FIG. 5 is in the range of 160 to 175, 325 to 340, and 510 to 525. The grayscale value of the recognition region to R3 and the surrounding grayscale There is a significant difference in the value, by finding the lowest gray level φ cloth position of the three identification areas R1 to 3, that is, the three corresponding features of the fingerprint of the single finger at positions 160, 326 and 515. The selection of the region; in addition, for example, the midline gray scale distribution shown in the lower end portion of FIG. 7 has the identification region R4 at 25 〇 to 275, and the lowest gray scale distribution position is also selected as the palm in the identification region R4. The selection of the corresponding feature area of the fingerprint, in fact, the selection of the feature area is the position of the middle main line among the three main lines closest to the palm. 'The change in the gray scale distribution is mainly due to the fact that there are 3 more obvious wrinkles (ie, deeper wrinkles) in a single finger, so the other parts of the finger will have grayscale changes, each finger There are 3 feature areas. Therefore, in this step, at least 15 feature areas can be selected in the finger fingerprint portion, and then the center point of the feature area is used, and 60×60 pixel points are set for the original hand fingerprint images. The feature area of "pixel" is "丨 to"] '. Similarly, as shown in Fig. 8, the center-point of the feature area of the palm print is set to 125 on the left and right sides, and the image is set to 125 pixels, and 75 pixels are set upward and 175 pixels are set downward to form A feature area M4 having 25 〇χ 25 〇 pixels, and thus as the identification range of the palm pattern, in order to obtain a more accurate identification effect, the feature area M4 can be the center point, ^10-1334578 • 96.10.31, 96131808 The patent specification, the 圚 type, and the patent scope modification are additionally selected from four feature regions M5 ′ and the four feature regions: respectively having 60×60 pixels. • Referring to FIGS. 1 and 4 to 9 again, the step (S4) of the fingerprint and palmprint dual feature identification method according to the preferred embodiment of the present invention mainly passes the selected number of patching regions M1 to M5 through the reward method. For example, the correlation coefficient method of the preferred embodiment of the present invention [c. lati 〇 〇 〇 ici ici ici 逐 逐 逐 逐 逐 与 与 与 与 与 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数 数The feature areas M1 to M5 may be compared with the preset scores set by the domain 3 and set by the default scores. If the comparison score is higher than the preset score, the comparison result is the feature region M1. The image to M5 conforms to the image of the corresponding template. Therefore, the fingerprint and palmprint dual feature recognition method of the present invention can simultaneously identify the plurality of feature regions M1 to M5 selected by the palm print and the hand fingerprint, and compare the identification methods to further improve the execution speed and Identify the effect. - As described above, compared with the conventional technology, the overall recognition time is long and the comparison time is long, and the image capturing unit 2 performs the palmprint and the hand fingerprint image along the set reference lines u and L2. Gray-scale processing; 'and borrowing (4) domain 3 riding ash _ change selects several feature areas M1 to M5 at the position of the palm print and the hand fingerprint; the feature area M1 to M5 • (4) touches the template established in the host 3 It is possible to output the comparison result, which can improve the execution speed and recognition effect. The present invention has been disclosed in the above-described preferred embodiments, and is not intended to limit the present invention, and those skilled in the art can, without departing from the spirit of the invention, 11-1334578 96.10.31, No. 96131808, the specification and drawings. The scope of the invention is to be construed as being limited to the scope of the invention. The scope of the invention is defined by the scope of the appended claims.
—12 — 96._第咖8。8細說明書、圏式及申請娜圍修正本 【囷式簡單說明】 第！圖：本發縣'用於較讀施狀減及 辨識方法之裝置組合立體圖。 叉特敛 辨==於較佳實施例之指紋及掌紋雙特徵 平开L3移圖之圭實施例之移動裝置帶動攝像機產生 法佳實施例之指紋及掌紋雙特徵辨識方 第5圖：本發明較佳實施例之指紋及掌紋雙 法之手指紋影像及其線段灰階分佈示意圖。 辨識方 法本發明較佳實施例之指紋及掌紋雙特徵辨識方 針子手指紋位置選取特徵區域之示意圖。° 第本發输佳實關之減及敎雙 法之手掌紋影像及其線段灰階分佈示意圖。 辨識方 法發明較佳實施例之指紋及掌紋雙特徵辨識方 針t手掌纹位置選取特徵區域之示意圖。 第9圖：本發明較佳實施例之指紋及 =特一騎心，另選取 【主要元件符號說明】 10 定位平台 20攝像機 1 暗房 2 影像_取單元 —13 — 1334578 96.10.31第96131808號專利說明書、圖式及申請專利範圍修正本 21 光源 3 主機 L1 基準線 Ml 特徵區域 M3 特徵區域 M5 特徵區域 R2 辨識區域 R4 辨識區域 22 移動裝置 31 影像擷取卡 L2 基準線 M2 特徵區域 M4 特徵區域 R1 辨識區域 R3 辨識區域—12 — 96._第咖8. 8 Detailed instructions, 圏 type and application for Na Wai correction [Simplified description of the 】] Figure: A perspective view of the combination of equipment used in the county's application for subtraction and identification. The invention relates to the fingerprint and the palmprint double feature swinging L3 shifting image of the preferred embodiment. The mobile device drives the camera to generate the fingerprint and the palm print double feature recognition method of the preferred embodiment. FIG. 5: The present invention A fingerprint and palmprint dual-hand fingerprint image of a preferred embodiment and a grayscale distribution diagram of the line segment thereof. Identification Method A schematic diagram of a fingerprint and palmprint dual feature recognition method in accordance with a preferred embodiment of the present invention. ° The first part of the transmission and the reduction of the image of the palm of the hand and the gray scale distribution of the line segment. Identification Method The fingerprint and palmprint dual feature recognition method of the preferred embodiment of the present invention is a schematic diagram of the selection of the feature area of the palmprint position. Figure 9: fingerprint and = one riding heart in the preferred embodiment of the present invention, and selecting [main component symbol description] 10 positioning platform 20 camera 1 darkroom 2 image_taking unit-13 1334578 96.10.31 Patent No. 96131808 , schema and patent application scope revision 21 Light source 3 Host L1 Baseline M1 Feature area M3 Feature area M5 Feature area R2 Identification area R4 Identification area 22 Mobile device 31 Image capture card L2 Baseline M2 Feature area M4 Feature area R1 Identification Area R3 identification area