TWI555507B - Electrocardiograph (ecg)-based identity identifying system - Google Patents
Electrocardiograph (ecg)-based identity identifying system Download PDFInfo
- Publication number
- TWI555507B TWI555507B TW103145025A TW103145025A TWI555507B TW I555507 B TWI555507 B TW I555507B TW 103145025 A TW103145025 A TW 103145025A TW 103145025 A TW103145025 A TW 103145025A TW I555507 B TWI555507 B TW I555507B
- Authority
- TW
- Taiwan
- Prior art keywords
- average
- average time
- difference
- average potential
- fuzzy
- Prior art date
Links
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Description
本發明係有關一種心電圖輔助之身分辨識系統，尤指一種兼具配合活體判別之指紋辨識裝置相當創新與配合模糊邏輯進行判別可提高辨識廣度和精確度之心電圖輔助之身分辨識系統。 The invention relates to an electrocardiogram-assisted identity recognition system, in particular to an electrocardiogram-assisted identity recognition system which is quite innovative with a fingerprint identification device and cooperates with fuzzy logic to improve the recognition breadth and accuracy.
傳統生物辨識技術至少有下列幾種：光學指紋識別技術：從光學發射裝置發射的光線，射到手指上再反射回機器以獲取數據，並對比資料庫看是否一致。光學識別只能到達皮膚的表皮層，而不能到達真皮層，而且受手指表面是否乾淨影響較大。 There are at least the following types of traditional biometrics: optical fingerprinting technology: Light emitted from an optical emitting device is reflected onto a finger and then reflected back to the machine to obtain data, and is compared to the database for consistency. Optical recognition can only reach the epidermis of the skin, not the dermis, and it is greatly affected by the cleanness of the finger surface.
電容指紋識別技術：電容感測器識別是安裝兩個具有一定間隔的兩個電容，利用指紋的凹凸，在手指滑過指紋檢測儀器時接通或斷開兩個電容的電流以檢測指紋資料。電容感測器對手指的乾淨要求比較高，而且感測器表面使用矽材料，比較容易損壞。 Capacitive fingerprint identification technology: Capacitive sensor identification is to install two capacitors with a certain interval. Using the bump of the fingerprint, the current of two capacitors is turned on or off when the finger slides over the fingerprint detecting instrument to detect the fingerprint data. Capacitive sensors have higher requirements for the cleanliness of the fingers, and the surface of the sensor is made of tantalum material, which is relatively easy to damage.
生物射頻指紋識別技術：以射頻感測器發射微量的射頻信號，穿透手指的表皮層獲取裡層的紋路以獲取信息。這種方法對手指的乾淨程度要求較低。 Bio-RF fingerprinting technology: A small amount of radio frequency signal is emitted by a radio frequency sensor, and the inner layer of the finger is penetrated to obtain the inner layer of the texture to obtain information. This method requires less cleanliness of the fingers.
上述幾種傳統指紋辨識成功率雖高，但無法保證使用者是否 為活體，或指紋是否為他人所盜用，也就是說，違法之徒有可能利用複製的指紋，通過傳統的指紋辨識裝置。 Although the success rate of the above several traditional fingerprint recognition is high, it cannot guarantee whether the user is Whether the living body, or the fingerprint is stolen by others, that is, the offender is likely to use the copied fingerprint to pass the traditional fingerprint identification device.
有鑑於此，必需研發出可解決上述習用缺點之技術。 In view of this, it is necessary to develop a technique that can solve the above disadvantages.
本發明之目的，在於提供一種心電圖輔助之身分辨識系統，其兼具配合活體判別之指紋辨識裝置相當創新與配合模糊邏輯進行判別可提高辨識廣度和精確度等優點。特別是，本發明所欲解決之問題係在於目前尚無心電圖輔助之身分辨識系統等問題。 The object of the present invention is to provide an electrocardiogram-assisted identity recognition system, which has the advantages of being quite innovative with the identification of a living body and the combination of fuzzy logic to improve the recognition breadth and accuracy. In particular, the problem to be solved by the present invention is that there is currently no problem such as an electrocardiogram-assisted identity recognition system.
解決上述問題之技術手段係提供一種心電圖輔助之身分辨識系統，其包括：一心電圖量測模組，係具有一第一手指接觸元件及一第二手指接觸元件；該第一手指接觸元件及該第二手指接觸元件係分別用以接觸一使用者之兩手之手指頭而量測出一心電圖訊號，該心電圖訊號包括至少一個完整之心電訊號週期，該心電訊號週期包含心電圖領域所定義之複數個波形轉折點P、Q、R、S、T；一處理部，係具有：一資料庫，係連結該心電圖量測模組，並用以擷取而儲存該心電圖訊號；一特徵值取得模組，係連結該資料庫，用以擷取該心電訊號週期的平均波形之該波形轉折點P、Q、R、S、T各點之平均電位及平均時間，並計算出各波形轉折點PQ、PR、QR、PS、RS、QS、ST、RT、QT與PT間之平均電位差及平均時間差，而取得三十個元素值，其係被定義為 ：平均電位VP、平均電位VQ、平均電位VR、平均電位VS、平均電位VT；平均時間TP、平均時間TQ、平均時間TR、平均時間TS、平均時間TT；平均電位差VPQ、平均電位差VPR、平均電位差VQR、平均電位差VPS、平均電位差VRS、平均電位差VQS、平均電位差VST、平均電位差VPT、平均電位差VQT、平均電位差VPT；平均時間差TPQ、平均時間差TPR、平均時間差TQR、平均時間差TPS、平均時間差TRS、平均時間差TQS、平均時間差TST、平均時間差TRT、平均時間差TQT、平均時間差TPT；前述之三十元素值係儲存回該資料庫中；一判別模組，係連結該資料庫，並內建三十個判別閥值，該三十個判別閥值係分別對應該三十元素值；藉此，當受測者之兩手之手指頭分別接觸該第一、該第二手指接觸元件，該心電圖量測模組測得其心電圖訊號並儲存至該資料庫，該特徵值取得模組擷取該心電圖訊號，並運算而取得該心電圖訊號之相對應的三十元素值，供該判別模組依序與該三十個判別閥值進行比對，當任一元素值不符合相對應之該判別閥值，即判別該受測者與該使用者非同一者；並當三十元素值全符合相對應之該判別閥值，即判別該受測者與該使用者為同一者。 The technical means for solving the above problem is to provide an electrocardiogram-assisted identity recognition system, comprising: an electrocardiography measurement module having a first finger contact component and a second finger contact component; the first finger contact component and the The second finger contact component is configured to measure an ECG signal by contacting a finger of a user's two hands. The ECG signal includes at least one complete ECG signal period, and the ECG signal period includes a defined ECG field. a plurality of waveform turning points P, Q, R, S, and T; a processing unit having: a database connected to the electrocardiographic measuring module for storing and storing the electrocardiogram signal; and a feature value obtaining module Linking the database to capture the average potential and average time of the waveform turning points P, Q, R, S, and T of the average waveform of the ECG cycle, and calculate the waveform turning points PQ, PR , QR, PS, RS, QS, ST, RT, QT and PT average potential difference and average time difference, and obtain thirty element values, which are defined as: average potential V P , average electricity Bit V Q , average potential V R , average potential V S , average potential V T ; average time T P , average time T Q , average time T R , average time T S , average time T T ; average potential difference V PQ , average Potential difference V PR , average potential difference V QR , average potential difference V PS , average potential difference V RS , average potential difference V QS , average potential difference V ST , average potential difference V PT , average potential difference V QT , average potential difference V PT ; average time difference T PQ , average Time difference T PR , average time difference T QR , average time difference T PS , average time difference T RS , average time difference T QS , average time difference T ST , average time difference T RT , average time difference T QT , average time difference T PT ; the aforementioned thirty element value The system stores the data back into the database; a discriminating module is connected to the database, and 30 thresholds are built in, and the thirty discriminating thresholds respectively correspond to the value of thirty elements; The finger of the two hands of the tester respectively contacts the first and second finger contact elements, and the electrocardiogram measurement module measures the ECG signal and stores it in the database, and the feature value is obtained. The module captures the ECG signal and calculates a corresponding thirty element value of the ECG signal for the comparison module to sequentially compare with the thirty determination thresholds, when any element value does not match Corresponding to the discriminating threshold, that is, determining that the subject is not the same as the user; and determining that the subject is the same as the user when the thirty element values all meet the corresponding discriminating threshold By.
本發明之上述目的與優點，不難從下述所選用實施例之詳細 說明與附圖中，獲得深入瞭解。 The above objects and advantages of the present invention are not difficult to be described in detail from the following selected embodiments. In the description and the drawings, you will get a deeper understanding.
茲以下列實施例並配合圖式詳細說明本發明於後： The invention will be described in detail in the following examples in conjunction with the drawings:
10‧‧‧心電圖量測模組 10‧‧‧ECG measurement module
11‧‧‧第一手指接觸元件 11‧‧‧First finger contact element
12‧‧‧第二手指接觸元件 12‧‧‧Second finger contact element
20‧‧‧處理部 20‧‧‧Processing Department
21‧‧‧資料庫 21‧‧‧Database
22‧‧‧特徵值取得模組 22‧‧‧Feature Value Acquisition Module
23‧‧‧判別模組 23‧‧‧Discrimination module
231‧‧‧判別閥值 231‧‧‧Determination threshold
232‧‧‧比對裝置 232‧‧‧ comparison device
80‧‧‧心電圖訊號 80‧‧‧ECG signal
81‧‧‧心電訊號週期 81‧‧‧ ECG signal cycle
90‧‧‧使用者 90‧‧‧Users
91‧‧‧手指頭 91‧‧‧ Fingers
P、Q、R、S、T‧‧‧波形轉折點 P, Q, R, S, T‧‧‧ waveform turning points
VP、VQ、VR、VS、VT‧‧‧平均電位 V P , V Q , V R , V S , V T ‧‧‧ average potential
TP、TQ、TR、TS、TT‧‧‧平均時間 T P , T Q , T R , T S , T T ‧‧‧ average time
VPQ、VPR、VQR、VPS、VRS、VQS、VST、VRT、VQT、VPT‧‧‧平均電位差 V PQ , V PR , V QR , V PS , V RS , V QS , V ST , V RT , V QT , V PT ‧ ‧ average potential difference
TPQ、TPR、TQR、TPS、TRS、TQS、TST、TRT、TQT、TPT‧‧‧平均時間差 T PQ , T PR , T QR , T PS , T RS , T QS , T ST , T RT , T QT , T PT ‧ ‧ average time difference
A1、A2‧‧‧圖形 A1, A2‧‧‧ graphics
m1、m2、m3、m4、m5、m6、m7、m8、m9、m10、m11、m12、m13、m14、m15‧‧‧點 M1, m2, m3, m4, m5, m6, m7, m8, m9, m10, m11, m12, m13, m14, m15‧‧
第一圖係本發明之示意圖 The first figure is a schematic view of the present invention
第二圖係本發明之心電圖訊號之示意圖 The second figure is a schematic diagram of the electrocardiogram signal of the present invention.
第三A圖係第二圖之心電訊號週期之平均電位之波形圖 The third A picture is the waveform of the average potential of the ECG signal period in the second figure.
第三B圖係第二圖之心電訊號週期之平均時間之波形圖 The third B picture is the waveform of the average time of the ECG period of the second picture
第四A、第四B及第四C圖係分別為本發明之第一歸屬函數、第二歸屬函數、輸出端之歸屬函數之模糊推論之示意圖 The fourth A, fourth B, and fourth C diagrams are schematic diagrams of the fuzzy inference of the first attribution function, the second attribution function, and the attribution function of the output end, respectively.
第五圖係本發明之一對七人之之比對之圖形 The fifth figure is a graphic of one of the seven pairs of the present invention.
第六圖係本發明之一對四十九人之比對之圖形 The sixth figure is a graphic of the comparison of forty-nine people of the present invention.
第七圖係第五圖之其中一圖形之放大示意圖 The seventh figure is an enlarged schematic view of one of the figures of the fifth figure
參閱第一、第二、第三A及第三B圖，本發明係為一種心電圖輔助之身分辨識系統，其包括：一心電圖量測模組10，係具有一第一手指接觸元件11及一第二手指接觸元件12；該第一手指接觸元件11及該第二手指接觸元件12係分別用以接觸一使用者90之兩手之手指頭91而量測出一心電圖訊號80，該心電圖訊號80包括至少一個完整之心電訊號週期81，該心電訊號週期81包含心電圖領域所定義之複數個波形轉折點P、Q、R、S、T；一處理部20，係具有： 一資料庫21，係連結該心電圖量測模組10，並用以擷取而儲存該心電圖訊號80；一特徵值取得模組22，係連結該資料庫21，用以擷取該心電訊號週期81的平均波形之該波形轉折點P、Q、R、S、T各點之平均電位及平均時間(參閱表1-1、1-2、1-3、2-1、2-2、2-3)，並計算出各波形轉折點PQ、PR、QR、PS、RS、QS、ST、RT、QT與PT間之平均電位差及平均時間差，而取得三十個元素值，其係被定義為：平均電位VP、平均電位VQ、平均電位VR、平均電位VS、平均電位VT；平均時間TP、平均時間TQ、平均時間TR、平均時間TS、平均時間TT；平均電位差VPQ、平均電位差VPR、平均電位差VQR、平均電位差VPS、平均電位差VRS、平均電位差VQS、平均電位差VST、平均電位差VRT、平均電位差VQT、平均電位差VPT；平均時間差TPQ、平均時間差TPR、平均時間差TQR、平均時間差TPS、平均時間差TRS、平均時間差TQS、平均時間差TST、平均時間差TRT、平均時間差TQT、平均時間差TPT；前述之三十元素值係儲存回該資料庫21中；一判別模組23，係連結資料庫21，並內建三十個判別閥值231，該三十個判別閥值231係分別對應該三十元素值；藉此，當受測者之兩手之手指頭91(例如左手姆指與右手姆指)分別接觸該第一、該第二手指接觸元件11與12，該心電圖量測模組10 測得其心電圖訊號80並儲存至該資料庫21，該特徵值取得模組22擷取該心電圖訊號80，並運算而取得該心電圖訊號80之相對應的三十元素值，供該判別模組23依序與該三十個判別閥值231進行比對，當任一元素值不符合相對應之該判別閥值231，即判別該受測者與該使用者非同一者；並當三十元素值全符合相對應之該判別閥值231，即判別該受測者與該使用者為同一者。 Referring to the first, second, third, and third B, the present invention is an electrocardiogram-assisted identification system, comprising: an electrocardiography measurement module 10 having a first finger contact element 11 and a The second finger contact element 12; the first finger contact element 11 and the second finger contact element 12 are respectively used for contacting the finger 91 of the two hands of a user 90 to measure an electrocardiogram signal 80, the electrocardiogram signal 80 Included in the at least one complete ECG signal period 81, the ECG signal period 81 includes a plurality of waveform inflection points P, Q, R, S, T defined in the field of electrocardiogram; a processing unit 20 having: a database 21, The ECG measurement module 10 is coupled to the device for capturing and storing the ECG signal 80. A feature value acquisition module 22 is coupled to the database 21 for capturing the average waveform of the ECG cycle 81. The average potential and average time of each point of the waveform turning point P, Q, R, S, T (refer to Table 1-1, 1-2, 1-3, 2-1, 2-2, 2-3), and calculate The average potential between each waveform turning point PQ, PR, QR, PS, RS, QS, ST, RT, QT and PT And the average time difference, acquires thirty element value, which is based is defined as: average electric potential V P, the average potential V Q, the average potential V R, the average potential V S, the average potential V T; average time T P, the average time T Q , average time T R , average time T S , average time T T ; average potential difference V PQ , average potential difference V PR , average potential difference V QR , average potential difference V PS , average potential difference V RS , average potential difference V QS , average potential difference V ST , average potential difference V RT , average potential difference V QT , average potential difference V PT ; average time difference T PQ , average time difference T PR , average time difference T QR , average time difference T PS , average time difference T RS , average time difference T QS , average time difference T ST , average time difference T RT , average time difference T QT , average time difference T PT ; the aforementioned thirty element values are stored back into the database 21; a discriminating module 23 is connected to the database 21 and built in thirty a discriminating threshold 231, the thirty discriminating thresholds 231 respectively corresponding to thirty element values; thereby, when the subject's two fingers of the finger 91 (such as the left hand and the right hand thumb) The electrocardiographic measurement module 10 detects the ECG signal 80 and stores it in the database 21, and the feature value acquisition module 22 captures the ECG signal 80. And calculating and obtaining the corresponding thirty-element value of the electrocardiogram signal 80, for the discriminating module 23 to sequentially compare with the thirty discriminating thresholds 231, when any element value does not meet the corresponding value The threshold value 231 is determined, that is, the subject is determined to be different from the user; and when the thirty element values all meet the corresponding determination threshold 231, it is determined that the subject is the same as the user.
實務上，該心電圖輔助之身分辨識系統可應用於各式需要門禁管制或電腦啟用控制之領域。 In practice, the ECG-assisted identity recognition system can be applied to various fields requiring access control or computer enabled control.
就實際心電圖領域而言，該每一使用者90之心電圖訊號80，皆可以示波器顯示而為連續的複數個心電訊號週期81組成，每一心電訊號週期81皆具有專屬該使用者90之三十元素值。 In the field of actual electrocardiogram, the ECG signal 80 of each user 90 can be displayed by the oscilloscope and is composed of a continuous plurality of ECG periods 81, and each ECG period 81 has 90 users exclusive to the user. Ten element values.
原則上本案擷取該單一個心電訊號週期81之三十元素值，即可用以進行辨識，然而，若取更多個心電訊號週期81，而將其三十元素值平均，則可改善雜訊斥拒力，提高辨識精密度。 In principle, the case extracts the element value of the single ECG signal period 81, which can be used for identification. However, if more ECG periods 81 are taken and the 30 element values are averaged, the improvement can be improved. Noise is repelled to improve recognition precision.
舉例來講，配合下表1-1、1-2及1-3進行說明，以本人坐姿實際量測後的第一元素值為70.8，設定正負20%為上下限(即係數為0.2)，則其判別閥值231即介於84.9至56.64之間，計算方式如下：70.8+70.8*0.2=84.9或70.8-70.8*0.2=56.64。 For example, with the following table 1-1, 1-2, and 1-3, the first element value measured by the actual sitting posture is 70.8, and the positive and negative 20% is set as the upper and lower limits (ie, the coefficient is 0.2). Then, the discriminating threshold 231 is between 84.9 and 56.64, and the calculation is as follows: 70.8+70.8*0.2=84.9 or 70.8-70.8*0.2=56.64.
其他二十九個元素值之判別閥值231可依此產生，恕不贅述，當然此係數也可改為0.3、0.5或其他數值，依設定者自行設定，一般介於0.2至0.5間，或經實測後調整。 The other twenty-nine element value discriminating threshold 231 can be generated accordingly, and will not be described here. Of course, this coefficient can also be changed to 0.3, 0.5 or other values, which are set by the setter, generally between 0.2 and 0.5, or Adjusted after actual measurement.
而當不同時間有一人使用門禁管制或電腦啟用控制時，只要 三十個元素值全部均符合該判別閥值231時，則系統判別為同一人，可開啟門禁；否則判定為不同人，不通過。 And when one person uses access control or computer enable control at different times, as long as When all the thirty element values are consistent with the discriminating threshold 231, the system determines that the same person can open the access control; otherwise, it is determined to be a different person and does not pass.
關於該判別模組23，也可修改為一模糊轉換裝置，其為雙輸入及單輸出之模糊系統，並包含由複數個第一、第二模糊歸屬函數構成之規則表，如下所述：平均電位VP及平均時間TP輸入後則輸出一第一模糊輸出值，其為X1(參閱下表3-1)；平均電位VQ及平均時間TQ輸入後則輸出一第二模糊輸出值，其為X2；平均電位VR及平均時間TR輸入後則輸出一第三模糊輸出值，其為X3；平均電位VS及平均時間TS輸入後則輸出一第四模糊輸出值，其為X4；平均電位VT及平均時間TT輸入後則輸出一第五模糊輸出值，其為X5；平均電位差VPQ及平均時間差TPQ輸入後則輸出一第六糢糊輸出值，其為X6；平均電位差VPR及平均時間差TPR輸入後則輸出一第七糢糊 輸出值，其為X7；平均電位差VQR及平均時間差TQR輸入後則輸出一第八糢糊輸出值，其為X8；平均電位差VPS及平均時間差TPS輸入後則輸出一第九糢糊輸出值，其為X9；平均電位差VRS及平均時間差TRS輸入後則輸出一第十糢糊輸出值，其為X10；平均電位差VQS及平均時間差TQS輸入後則輸出一第十一糢糊輸出值，其為X11(參閱下表3-2)；平均電位差VST及平均時間差TST輸入後則輸出一第十二糢糊輸出值，其為X12；平均電位差VRT及平均時間差TRT輸入後則輸出一第十三糢糊輸出值，其為X13；平均電位差VQT及平均時間差TQT輸入後則輸出一第十四糢糊輸出值，其為X14；平均電位差VPT及平均時間差TPT輸入後則輸出一第十五糢糊輸出值，其為X15；X=[x 1 x 2......x 15]；另設一容許範圍A；A=[α 1 α 2......α 15]；其中α 1至α 15係為第一容許值至第十五容許值；X及A均存於該資料庫21中； 並該判別模組23為模糊控制轉換裝置時，又包括一比對裝置232：其係連結並由該心電圖量測模組10讀入一新的心電圖訊號80，並透過該特徵值取得模組22而計算出一組新的三十元素值；之後，對該比對裝置232輸入該新的三十元素值後得到一組新的規則表，其包括新的十五個糢糊輸出值Y1至Y15；若Y落在X±A之中，則判定為「身分辨識通過」；否則判定為「身分辨識不通過」。 The discriminating module 23 can also be modified into a fuzzy converting device, which is a two-input and single-output fuzzy system, and includes a rule table composed of a plurality of first and second fuzzy attribution functions, as follows: After the potential V P and the average time T P are input, a first fuzzy output value is output, which is X1 (refer to Table 3-1 below); after the average potential V Q and the average time T Q are input, a second fuzzy output value is output. , which is X2; the average potential V R and the average time T R are input to output a third fuzzy output value, which is X3; the average potential V S and the average time T S are input to output a fourth fuzzy output value, It is X4; after inputting the average potential V T and the average time T T , a fifth fuzzy output value is output, which is X5; after the average potential difference V PQ and the average time difference T PQ are input, a sixth fuzzy output value is output, which is X6. The average potential difference V PR and the average time difference T PR input a seventh fuzzy output value, which is X7; the average potential difference V QR and the average time difference T QR input then output an eighth fuzzy output value, which is X8; Potential difference V PS and average time difference T PS input Then, a ninth fuzzy output value is output, which is X9; after the average potential difference V RS and the average time difference T RS are input, a tenth fuzzy output value is output, which is X10; after the average potential difference V QS and the average time difference T QS are input, An eleventh fuzzy output value is output, which is X11 (refer to Table 3-2 below); after the average potential difference V ST and the average time difference T ST are input, a twelfth fuzzy output value is output, which is X12; the average potential difference V RT And after the average time difference T RT input, a thirteenth fuzzy output value is output, which is X13; after the average potential difference V QT and the average time difference T QT are input, a fourteenth fuzzy output value is output, which is X14; the average potential difference V PT And after the average time difference T PT is input, a fifteenth fuzzy output value is output, which is X15; X = [ x 1 x 2 ... x 15 ]; another allowable range A; A = [ α 1 α 2 ...... α 15 ]; wherein α 1 to α 15 are the first allowable value to the fifteenth allowable value; X and A are both stored in the database 21; and the discriminating module 23 is When the control device is fuzzy controlled, it further includes a matching device 232: the system is connected and the new reading is read by the electrocardiography measuring module 10 The signal signal 80 is calculated by the feature value obtaining module 22, and a new set of thirty element values is calculated. Then, the new thirty-element value is input to the comparing device 232 to obtain a new set of rule tables. It includes new fifteen fuzzy output values Y1 to Y15; if Y falls within X±A, it is judged as "identification identification passed"; otherwise, it is judged as "identification identification does not pass".
關於下表3-1及表3-2，其係以「使用者坐」(即使用者坐姿)之資料為參考基準，與其比較後之差異。例如：受測者甲之第一模糊輸出值與使用者坐姿時之第一模糊輸出值相減後之結果為-17.37，即以(17.37)表示，而正負之標準值為16.21及-16.21，由於-17.37之絕對值大於16.21，所以算是『不通過』。 For the following table 3-1 and Table 3-2, the data of the "user sitting" (ie, the user's sitting position) is used as a reference and the difference is compared. For example, the result of subtracting the first fuzzy output value of the subject A from the first fuzzy output value when the user is sitting is -17.37, which is represented by (17.37), and the standard values of positive and negative are 16.21 and -16.21, Since the absolute value of -17.37 is greater than 16.21, it is considered "not passed."
關於模糊系統，更詳細的講，參閱下表4-1、第四A、第四B及第四C及第五圖：表4-1
第一輸入歸屬函數包含P1、P2、P3、Z、N1、N2、N3；第二輸入輸歸屬函數包含n1、n2、n3、z、p1、p2、p3；輸出端之歸屬函數包含Po1、Po2、Po3、Zo、No1、No2、No3。 The first input attribution function includes P 1 , P 2 , P 3 , Z, N 1 , N 2 , N 3 ; the second input-delivery function includes n 1 , n 2 , n 3 , z, p 1 , p 2 , p 3 ; the attribution function of the output includes Po 1 , Po 2 , Po 3 , Zo, No 1 , No 2 , No 3 .
以上歸屬函數的數目可根據需求選用更多(得到較精確的辨識結果)或更少(得到較粗略但可縮短運算時間的辨識結果)個歸屬函數。因演算過程類似恕不贅述。 The number of the above-mentioned attribution functions may be selected according to requirements (more accurate identification results) or less (to obtain a more rough but shortened operation time identification result) belonging functions. Because the calculation process is similar, I will not go into details.
此外，參閱第五圖，其係為甲、乙、丙、丁、戊、己、「使用者坐」及「使用者站」(依序從上排左至上排右，再從下排左至下排右)共8組，之其與「使用者坐」相減後之第一至第十五模糊輸出值，這15個值，以一點為圓心，每隔24度(即360/15)向外匯出一點，即可形成一差異特徵圖(或稱雷達圖)，由15個點(即第七圖中標示之點m1至m15)所構成之多邊形，可方便以視覺來判斷差異大小。 In addition, please refer to the fifth figure, which is A, B, C, D, E, H, "Sit" and "User Station" (sequentially from the top left to the top, then from the bottom left to the left) The lower row is right) A total of 8 groups, which are the first to fifteenth fuzzy output values after subtraction from the "user sitting". These 15 values are centered at one point, every 24 degrees (ie 360/15). If you make a point to the foreign exchange, you can form a difference feature map (or radar chart). The polygon formed by 15 points (the points m1 to m15 marked in the seventh figure) can be used to visually judge the difference size.
而這8個圖中，由於第7筆為「使用者坐」與「使用者坐」相減之結果，將成為一點(標為A1)。而第一至第六個圖所示之圖形顯示有 相當之差異，而第八個圖之差異相對較小。 In the eight figures, since the seventh paragraph is the result of the subtraction between "user sitting" and "user sitting", it will become a point (labeled A1). The figures shown in the first to sixth figures show The difference is quite different, and the difference between the eighth figures is relatively small.
再者，如第六圖所示，共有50筆資料，分別代表50個受測者，其中之第七筆為「使用者坐」(使用者自己)。此50筆資料係為與第七筆為「使用者坐」相減後之差值，其中圖形A2為使用者之資料，明顯與其他四十九個不同，足以辨識，且可看出每個人都有自已獨特之圖形。 Furthermore, as shown in the sixth figure, there are 50 pieces of data representing 50 subjects, the seventh of which is "user sitting" (users themselves). The 50 data is the difference between the seventh and the "user sitting". The graphic A2 is the user's data, which is obviously different from the other forty-nine. It is enough to identify and everyone can be seen. Have their own unique graphics.
綜上所述，本創作之優點及功效可歸納為： In summary, the advantages and effects of this creation can be summarized as:
[1]可配合活體判別之指紋辨識裝置相當創新。傳統的指紋辨識裝置，並沒有活體的判別機能，也就是說，違法之徒有可能利用複製出來的指紋，即有可能通過傳統的指紋辨識裝置；而本創作以心電圖(Electrocardiogram，簡稱ECG)之量測來進行輔助，配合心電圖特徵之模糊化辨識機制可有效提高安全性。故，可配合活體判別之指紋辨識裝置相當創新。 [1] The fingerprint identification device that can be used in conjunction with living body identification is quite innovative. The traditional fingerprint identification device does not have the function of discriminating in vivo. That is to say, it is possible for the offender to use the copied fingerprint, that is, it is possible to pass the traditional fingerprint identification device; and the creation is based on an electrocardiogram (ECG). Measurement to assist, with the fuzzy identification mechanism of ECG features can effectively improve safety. Therefore, the fingerprint identification device that can cooperate with the living body discrimination is quite innovative.
[2]配合模糊邏輯進行判別可提高辨識廣度和精確度。本發明以模糊邏輯設計人體心電圖為基礎的身分辨識系統，可以適用人體心電圖訊號因為姿態或體能狀態不同而致太大的變異。因為模糊邏輯具有較佳的強健性，所以適用ECG辨識系統變異較大的問題。而且可隨時增加外來考量因素，也就是增加輸入研判因素的數目以提昇辨識的廣度和精準度。故，配合模糊邏輯進行判別可提高辨識的廣度和精確度。 [2] Discrimination with fuzzy logic can improve the breadth and accuracy of identification. The invention adopts fuzzy logic to design a body identification system based on the human body electrocardiogram, and can be applied to the human body electrocardiogram signal because the posture or the physical state is different and the variation is too large. Because fuzzy logic has better robustness, it is suitable for the problem of large variation of ECG identification system. Moreover, external factors can be increased at any time, that is, the number of input factors is increased to increase the breadth and accuracy of the identification. Therefore, the discrimination with fuzzy logic can improve the breadth and accuracy of recognition.
以上僅是藉由較佳實施例詳細說明本發明，對於該實施例所做的任何簡單修改與變化，皆不脫離本發明之精神與範圍。 The present invention has been described in detail with reference to the preferred embodiments of the present invention, without departing from the spirit and scope of the invention.
10‧‧‧心電圖量測模組 10‧‧‧ECG measurement module
11‧‧‧第一手指接觸元件 11‧‧‧First finger contact element
12‧‧‧第二手指接觸元件 12‧‧‧Second finger contact element
20‧‧‧處理部 20‧‧‧Processing Department
21‧‧‧資料庫 21‧‧‧Database
22‧‧‧特徵值取得模組 22‧‧‧Feature Value Acquisition Module
23‧‧‧判別模組 23‧‧‧Discrimination module
231‧‧‧判別閥值 231‧‧‧Determination threshold
232‧‧‧比對裝置 232‧‧‧ comparison device
90‧‧‧使用者 90‧‧‧Users
91‧‧‧手指頭 91‧‧‧ Fingers
Claims (1)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW103145025A TWI555507B (en) | 2014-12-23 | 2014-12-23 | Electrocardiograph (ecg)-based identity identifying system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW103145025A TWI555507B (en) | 2014-12-23 | 2014-12-23 | Electrocardiograph (ecg)-based identity identifying system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201622645A TW201622645A (en) | 2016-07-01 |
| TWI555507B true TWI555507B (en) | 2016-11-01 |
Family
ID=56984386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW103145025A TWI555507B (en) | 2014-12-23 | 2014-12-23 | Electrocardiograph (ecg)-based identity identifying system |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI555507B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10238301B2 (en) * | 2016-11-15 | 2019-03-26 | Avidhrt, Inc. | Vital monitoring device, system, and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101773394A (en) * | 2010-01-06 | 2010-07-14 | 中国航天员科研训练中心 | Identification method and identification system using identification method |
| TWM453912U (en) * | 2012-12-05 | 2013-05-21 | Nat Univ Chung Hsing | Fingerprint recognition device with electrocardiogram assistance |
-
2014
- 2014-12-23 TW TW103145025A patent/TWI555507B/en not_active IP Right Cessation
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101773394A (en) * | 2010-01-06 | 2010-07-14 | 中国航天员科研训练中心 | Identification method and identification system using identification method |
| TWM453912U (en) * | 2012-12-05 | 2013-05-21 | Nat Univ Chung Hsing | Fingerprint recognition device with electrocardiogram assistance |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201622645A (en) | 2016-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nguyen et al. | Robust biometric recognition from palm depth images for gloved hands | |
| EP3158501A1 (en) | Method and apparatus for biometric-based security using capacitive profiles | |
| US9851834B2 (en) | Time domain differential techniques to characterize various stimuli | |
| Nabila et al. | Gait‐based human age classification using a silhouette model | |
| Kim et al. | 1D CNN based human respiration pattern recognition using ultra wideband radar | |
| CN103315744A (en) | Hand tremor detection method | |
| KR102570385B1 (en) | A method and electronic device for determiing whether to allow user access | |
| Ghaderyan et al. | Neurodegenerative diseases detection using distance metrics and sparse coding: A new perspective on gait symmetric features | |
| CN108875629A (en) | Vena metacarpea recognition methods based on multisample Fusion Features | |
| Gao et al. | Expression robust 3D face landmarking using thresholded surface normals | |
| TWI555507B (en) | Electrocardiograph (ecg)-based identity identifying system | |
| Kang et al. | A Precise Muscle activity onset/offset detection via EMG signal | |
| CN102622039A (en) | Intelligent control handle based on natural contact sensors and applications thereof | |
| Mancilla-Palestina et al. | Embedded system for bimodal biometrics with fiducial feature extraction on ecg and ppg signals | |
| Stefanou et al. | Tactile signatures and hand motion intent recognition for wearable assistive devices | |
| Xie et al. | Fingerprint quality analysis and estimation for fingerprint matching | |
| Zeng et al. | Accelerometer-based gait recognition via deterministic learning | |
| Liu et al. | A respiration-derived posture method based on dual-channel respiration impedance signals | |
| KR20160035497A (en) | Body analysis system based on motion analysis using skeleton information | |
| CN111436940A (en) | Gait health assessment method and device | |
| Cao et al. | Structure feature extraction for finger-vein recognition | |
| Ferdinando et al. | Bivariate empirical mode decomposition for ECG-based biometric identification with emotional data | |
| Likitlersuang et al. | Arm angle detection in egocentric video of upper extremity tasks | |
| Krishnan et al. | Abnormal gait detection using lean and ramp angle features | |
| Peiqin et al. | Finger vein recognition algorithm based on optimized GHT |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MM4A | Annulment or lapse of patent due to non-payment of fees |