Fingerprint identification enhancement method

Abstract

The invention provides a fingerprint identification enhancement method, and relates to the technical field of fingerprint identification, and the method comprises the steps: collecting the fingerprint data of a user through a fingerprint sensor, generating a multi-dimensional operation data set, extracting at least one piece of extra dimension information based on the multi-dimensional operation data set, and carrying out the verification of at least one piece of extra dimension information according to a pre-stored verification rule. Performing combined verification on the fingerprint data and the additional dimension information, and if a combined verification result is matched with a pre-stored standard, authorizing user access; otherwise, refusing the access. Multi-factor authentication of'biological characteristics + behavior characteristics' is realized by fusing multi-dimensional operation information such as direction and time with fingerprint characteristics, a dynamic behavior model is constructed by using sensor data on the premise of not increasing hardware cost, anti-copying and anti-cheating capabilities are remarkably improved, user operation is simplified through intelligent UI guidance, and user experience is improved. The contradiction between safety and usability is effectively cracked, the overall safety of the system can be greatly improved at the minimum operation cost, and the method is particularly suitable for a high-safety requirement scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fingerprint identification, in particular to a fingerprint identification enhancement method. BACKGROUND

[0002] The requirement of fingerprint identification is relatively high, and various situations affect fingerprint identification, for example: 1. The humidity, cleanliness, grease, stains and the like of the finger; 2. Few fingerprint features or more wear and tear, etc. 3. Skin damage or scarring, etc.

[0003] In addition, there is a risk of fingerprint copying in fingerprint identification. Each time the fingerprint is identified, the user's fingerprint will be left on the fingerprint collection head, which can be used for fingerprint copying, and there is a conflict between security and ease of use.

[0004] It should be noted that there is a certain degree of mutual conflict between biometric technology in security and ease of use. The false acceptance rate (FAR) is the proportion of the system that incorrectly accepts illegal users. The higher the FAR, the less secure it is, and vice versa. The false rejection rate (FRR) is the proportion of the system that incorrectly rejects legal users. The higher the FRR, the worse the user experience, and vice versa. FAR and FRR usually cancel each other out and cannot be reduced indefinitely at the same time. Therefore, the common classification (consumer / commercial / military) of biometric technology only improves a few levels in the FAR and FRR indicators, and there is no way to improve the security and accuracy like a password lock. Therefore, the present application proposes several ways to improve security and accuracy at the minimum cost of user experience based on existing technology. In summary, it is necessary to design a fingerprint identification enhancement method. SUMMARY

[0005] In order to overcome the shortcomings of the prior art, the purpose of the present application is to provide a fingerprint identification enhancement method.

[0006] To achieve the above purpose, the present application provides the following solutions: The present application provides a fingerprint identification enhancement method, comprising: Step 1: collecting the fingerprint data of the user through the fingerprint sensor and generating a multi-dimensional operation data set; Step 2: extracting at least one additional dimension information based on the multi-dimensional operation data set; Step 3: combining and verifying the fingerprint data and the additional dimension information according to the pre-stored verification rule; Step 4: if the combined verification result matches the pre-stored standard, the user is authorized to access; otherwise, access is denied.

[0007] Preferably, the multidimensional operational dataset includes fingerprint image data, coordinate location data, and time series data.

[0008] Preferably, the coordinate position data is obtained by two-dimensional coordinate acquisition using a fingerprint sensor.

[0009] Preferably, the additional dimension signal includes direction dimension information, time dimension information, or quantity / area dimension information.

[0010] Preferably, the directional dimension information is determined by analyzing changes in coordinate position data to determine the coordinate direction flow.

[0011] Preferably, the coordinate direction flow is determined by analyzing changes in coordinate position data, specifically as follows: Multiple frames of coordinate data are continuously collected using a fingerprint sensor; Identify the type of data in each frame; The coordinate direction flow is determined based on the type change of adjacent frames.

[0012] Preferably, the coordinate direction flow is determined by analyzing changes in coordinate position data, specifically as follows: Multiple frames of coordinate data are continuously collected using a fingerprint sensor; The coordinate direction flow is determined based on the time series changes of multi-needle coordinate data.

[0013] Preferably, the time dimension information is determined by calculating the touch duration of the fingerprint sensor, including long touch, short touch, or a specific time series.

[0014] Preferably, the quantity / area dimension information is determined by identifying the effective touch area or combined pattern of the fingerprint sensor.

[0015] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: This invention provides a fingerprint recognition enhancement method. The method includes collecting a user's fingerprint data using a fingerprint sensor and generating a multi-dimensional operation dataset. Based on the multi-dimensional operation dataset, at least one additional dimension of information is extracted. According to pre-stored verification rules, the fingerprint data and the additional dimension information are combined for verification. If the combined verification result matches the pre-stored standard, user access is authorized; otherwise, access is denied. This invention achieves the core effect of maximizing security with minimal user operation cost by introducing multi-dimensional operation information such as direction, time, and area into fingerprint biometric verification. Specifically, without changing the existing fingerprint sensor hardware architecture, this method innovatively utilizes the coordinate and time sequences collected by the sensor to construct dynamic behavioral features such as "coordinate direction flow" and "operation rhythm," upgrading the traditional single static fingerprint image comparison to multi-factor authentication of "biometric features + behavioral features." This significantly enhances the system's anti-copying and anti-spoofing capabilities, because even if fingerprint information is stolen, attackers find it difficult to imitate the user's specific sliding direction, rhythm, and trajectory combination. Simultaneously, this method reduces the user's learning cost through intelligent guided UI, transforming complex security logic into intuitive sliding operations, effectively resolving the inherent contradiction between security and ease of use. Ultimately, the system significantly reduces the false acceptance rate and improves reliability in high-security scenarios with almost no increase in verification time. It also provides a more robust verification pass rate for users with blurred or worn fingerprint features, achieving the best balance between security, experience, and cost. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of EMPTY type data; Figure 3 This is a schematic diagram of FULL type data; Figure 4 This is a diagram similar to the data; where, Figure 4 'a' is a schematic diagram of type A data. Figure 4 b is a schematic diagram of type a data. Figure 4 c is a schematic diagram of type D data. Figure 4 d is a schematic diagram of d-type data; Figure 5 This is a diagram illustrating the subclass data of type A.

[0018] Figure 6 A schematic diagram illustrating the UI design for a circular shape. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The purpose of this invention is to provide a fingerprint recognition enhancement method that integrates multi-dimensional operational information such as orientation and time with fingerprint features to achieve multi-factor authentication combining biometrics and behavioral characteristics. This method significantly improves anti-copying and anti-spoofing capabilities by constructing a dynamic behavioral model using sensor data without increasing hardware costs. The intelligent UI guides users through simplified operations, effectively resolving the conflict between security and ease of use. It can significantly improve overall system security with minimal operational overhead, making it particularly suitable for high-security scenarios.

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] First, the basic principles of this invention will be introduced. This invention utilizes specific value combinations to form an independent recognition logic, specifically combining fingerprint recognition with specific combination logic to form a complete fingerprint recognition scheme.

[0023] Specifically, there are three schemes based on the following principles: 1. Fingerprint sensor directional dimension combination method: Touching a fingerprint sensor can add directional attributes, thus adding at least 8 specific values: ↑, ↖, ←, ↙, ↓, ↘, →, ↗.

[0024] 2. Fingerprint sensor time dimension combination method: The touch of a fingerprint sensor can be supplemented with a time attribute, thus adding at least two specific values: long and short. To put it simply, it is essentially adding a counter logic, taking 1 second or 2 seconds as an example, which can be implemented through sensor touch and UI design. For example, if the sensor is used as a button, the time interval from pressing to releasing is visible and can be combined to form a password, thereby realizing related functions.

[0025] 3. Combination method of fingerprint sensor quantity / area dimension; This can be achieved by increasing the area of ​​the fingerprint sensor or by combining multiple sensors into a single array. It's even possible to create specific pattern values. Multiple sensors can be combined for this purpose; the logic is relatively simple and will not be described in detail here.

[0026] Taking 1 as an example, under the condition of adding specific input conditions, the verification process can be based on fingerprint recognition, while requiring 3 rounds of verification in the following directions: ↙, ↙, →.

[0027] Taking 2 as an example, under the condition of supplementing specific input conditions, the verification process can be based on fingerprint recognition, while requiring 3 rounds of verification in the following directions: short, long, short.

[0028] In scenarios where a specific graphical UI is used, more detailed and refined feature points and verification conditions can be created.

[0029] First, the present invention provides a fingerprint recognition enhancement system, including a fingerprint sensor, a computing unit (CPU / MCU), and a boot UI; The fingerprint sensor, a data acquisition unit, is responsible for collecting fingerprints; CPU / MCU, the computing unit, is responsible for comparing and verifying fingerprint feature points. As a logic control unit, it connects the upper and lower layers and is responsible for the intermediate processing of logic control of the sensor and guidance of the UI, etc. The UI guide, the external display, is responsible for guiding users through all processes, including but not limited to physical hardware interfaces, software interfaces, voice prompts, lighting effects, etc.

[0030] This invention innovatively expands the sampling range of fingerprint sensors, no longer limited to valid images, and uses them as reliable information to enhance and optimize the certainty that conventional fingerprint schemes cannot achieve 100%.

[0031] The method will then be described in detail based on the system: like Figure 1 As shown, the present invention provides a fingerprint recognition enhancement method, comprising: Step 1: Collect the user's fingerprint data using a fingerprint sensor and generate a multidimensional operation dataset; Step 2: Based on the multidimensional operation dataset, extract at least one additional dimension of information; Step 3: Combine and verify the fingerprint data with additional dimension information according to the pre-stored verification rules; Step 4: If the combined verification result matches the pre-stored criteria, then authorize the user to access; otherwise, deny access.

[0032] In step 1, the user's fingerprint data is collected using a fingerprint sensor, and a multidimensional operational dataset is generated, specifically as follows: The multidimensional operational dataset includes fingerprint image data, coordinate location data, and time series data; This article introduces the classification of common fingerprint sensors and their imaging principles: Currently, the mainstream fingerprint sensors on the market mainly include three types: capacitive, optical, and ultrasonic. Their core objective is to transform the "ridges" and "valleys" of a fingerprint into a digital image. Capacitive sensors (ridges - high capacitance, valleys - low capacitance), optical sensors (ridges - total internal reflection - bright areas, valleys - diffuse reflection - dark areas), and ultrasonic sensors (ridges - strong reflection - bright areas, valleys - weak reflection - dark areas). Regardless of the type, based on their imaging timing and image transition patterns, they create possibilities for orientation / time, etc. Currently, the main shapes of fingerprint sensors and devices on the market are circular and rectangular.

[0033] Information collected by a fingerprint sensor: The fingerprint sensor's acquisition surface can be represented by 2D coordinates, with the coordinate axis range depending on the sensor's area and accuracy. A fingerprint sensor can collect information at any time and digitize it. Digitized information can be defined as: 1 - data present, 0 - no data, specifically as follows... Figure 2 As shown; like Figure 2 As shown, in the entire 2D coordinate digital table, there are many blocks on all x and y axes with a digital information of 0, that is, there is no object on the sensor (including fingers). like Figure 3 As shown, if a large amount of data is located in the center of most of the sensor area, the CPU / MCU can obtain a sufficient amount of contrast metadata when extracting feature values. Therefore, when the data presents this feature, it is considered a valid candidate for imaging. Note: In reality, the sensor area and the corresponding 2D coordinates are definitely not as large as (4x4). This is just a simple analogy for easy understanding.

[0034] like Figure 4 As shown, when only the edges of the coordinate system contain data, this data, even if it comes from a finger, cannot be used because there are not enough fingerprint feature values. Here, based on its data sampling characteristics, it is defined as data of class A(bc), a(BC), D(bc), d(BC)... It is obvious that Figure 5 The form of information collected on one side of the fingerprint sensor can also be classified as type A (bc) (or as a subclass of it).

[0035] In step 2, based on the multidimensional operation dataset, at least one additional dimension of information is extracted, specifically: The additional dimension signal includes orientation dimension information, time dimension information, or quantity / area dimension information; This section uses directional dimension information as an example for explanation: CPU / MCU performs feature classification and coordinate direction flow extraction for multidimensional operation datasets. The CPU / MCU can define the data as EMPTY type based on the above data characteristics. Figure 2 ), FULL type (True / False) ( Figure 3 Type A Figure 4 -a), type a ( Figure 4 -b), D type ( Figure 4 -c), d type ( Figure 4 -d) ...and several others; EMPTY type ( Figure 2 There is no object on the sensor; FULL type (True / False) ( Figure 3 The sensor collected most of the object data (there is no need to determine whether it is fingerprint information at this time). The remaining types all have data on one side of the sensor, but no complete and valid data. Older schemes would typically discard this type of data, while new schemes can collect and organize it as needed. For CPUs / MCUs, there are no directional features such as ↑↓←→ (the fingerprint orientation field is defined relative to fingerprint features), but there can be coordinate direction flow.

[0036] The CPU / MCU reads the data sampled multiple times in chronological order, and the following can be seen: Case 1: Empty type... D type... FULL type... A type... Empty type; Case 2: Empty type...Full type...Type A...Empty type; Case 3: Empty type...Full type...Empty type; Taking the extraction of coordinate direction flow as an example, there are two methods: (1) The simplest design is that the CPU / MCU only needs to record two adjacent different types within the sampling period (recording multiple times only complicates the application and does not increase the accuracy) to obtain the coordinate direction flow. Taking Case 1 or Case 2 as an example, the coordinate direction flow is A out (A type >>> Empty type). Similarly, we can derive various types: a out, D out, d out, etc.; In addition, there can be more complex directional flow definitions. For example, Case 1 above is a complex coordinate directional flow: D in, A out, etc. can be used as extended design directions. The principle is similar, and it will not be elaborated here. (2) Another design method for obtaining coordinate direction flow: When there is an object (finger), the sensor sampling time will be a fixed value (because the reflection distance is fixed), while when there is no finger, the sampling time will time out. Organizing the sampling time according to the sensor plane coordinate axis will result in a data time table similar to a data storage table: 1 - quickly obtain information 0 - timeout. Due to the similar design, it will not be elaborated in detail here.

[0037] The onboarding UI also needs to be designed, such as Figure 6 As shown, taking a circular shape as an example, the sensor can be divided into several areas (e.g., 16 areas) through software / hardware / mold design. (1) When the finger is deliberately swiped from bottom to top, the sensor will first sample and image in the 8 / 9 region (fingerprint invalid), and then finally sample and image in the 16 / 1 region (fingerprint invalid). If such a pattern is presented as (8,9) in>>> (16,1) out, we can define it as ↑; (2) Furthermore, when the fingerprint sampling ends (the fingerprint is valid), the fingerprint moves upwards, that is, it moves directly out at (16, 1), which can be defined as ↑; other examples such as (2, 3) out--↗, (4, 5) out--→... and so on, which will not be elaborated here; (3) Furthermore, without fingerprint imaging, the finger sliding direction can also be determined by the time difference of the capacitance / light wave signal (the change trajectory of reflection >>> no reflection is the direction). A rectangular sensor can also be divided into regions in a similar way to determine the direction of finger swipes. Mold appearance or UI animation design can make the changes more precise. Associating and binding sensor coordinate direction flow and UI design direction flow: linking UI graphics ( Figure 6 ) and sensor coordinate graph ( Figures 2-5 By overlapping, the binding relationship can be easily determined. In the design, the direction of the sensor coordinate graph is not important; it can be adjusted regardless of its placement. If the sensor and UI (mold / software UI / hardware UI) are placed according to the directions shown in the diagram, it can be concluded that the coordinate direction flow A is equivalent to ↑; other directions can be analogized and will not be elaborated further.

[0038] In addition, the time dimension information is determined by calculating the touch duration of the fingerprint sensor, including long touch, short touch, or specific time series (similar to the design of the coordinate direction flow dimension can be easily achieved by using the CPU / MCU timer + the sensor reflection time data mentioned above + UI). The quantity / area dimension information (such as the touch sequence of multiple fingerprint sensors, which can also be used as an independent dimension, or even a more complex combination of coordinate direction flow and time data image combination of multiple sensors) is determined by identifying the effective touch area or combination pattern of the fingerprint sensor.

[0039] This invention provides an embodiment to illustrate the above method through a practical example, specifically as follows: 1. A cipher combination composed of one direction.

[0040] Password registration -- Direction (←→←→) Pre-enter system >>> Login mode 1 >>> Swipe your finger in a specific direction (←→←→) >>> Match successful >>> Obtain authorization; 2. A password combination consisting of fingerprint and direction.

[0041] Password registration -- Direction (←→←→) pre-enrollment system >>> Fingerprint pre-enrollment system >>> Login mode 2 >>> Swipe your finger in a specific direction ((fingerprint ←)(fingerprint →)(fingerprint ←)(fingerprint →)) >>> (fingerprint + direction) match successful >>> Obtain authorization; 3. A password combination consisting of fingerprint, direction, and time.

[0042] Password 1 Registration -- Direction (←→←→) Pre-enter system >>> Password 2 Registration -- Time (Long Long Short Long) Pre-enter system >>> Fingerprint Pre-enter system >>> Login Mode 3 >>> Swipe your finger in a specific direction ((Fingerprint + Long + ←)(Fingerprint + Long + →)(Fingerprint + Short + ←)(Fingerprint + Long + →)) >>> (Fingerprint + Direction + Time) Match successful >>> Obtain authorization; Obviously, the dimensions can be arranged in any combination and order according to the design; the definition of time, combined with the UI design, can be upgraded to specific time (3 seconds, 2 seconds, 5 seconds, 2 seconds, etc.).

[0043] Taking a unidirectional cryptographic combination as an example, the specific steps are explained as follows: 1. The CPU / MCU pre-stores the logic UI direction data as a "password"; The CPU / MCU guides the user to perform operations through the UI, while simultaneously controlling the sensor to collect multiple rounds of data; The CPU / MCU analyzes, organizes, and compares the sampled data, and "translates" the coordinate direction flow to the logical UI direction; The CPU / MCU compares the pre-stored data; if they match, the match is successful; otherwise, it fails.

[0044] 2. The CPU / MCU translates the logical UI direction data entered by the user into coordinate direction correction data and stores it as a "password" in advance; The CPU / MCU guides the user to perform operations through the UI, while simultaneously controlling the sensor to collect multiple rounds of data; The CPU / MCU compares the newly acquired coordinate direction data with the pre-stored "password". If they match, the match is successful; otherwise, it fails.

[0045] It can be seen that the main difference between processes 1 and 2 lies in the data type being compared between the CPU / MCU: whether it's logical UI-oriented data or coordinate-oriented flow data. In essence, there is no difference. The password combinations consisting of fingerprint + direction and fingerprint + direction + time are merely enhancements and complication of the guidance UI and process design, without any fundamental difference, and will not be listed or elaborated here.

[0046] It should also be noted that time-related information can be relatively easily implemented using the timer function of the CPU / MCU.

[0047] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0048] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A fingerprint recognition enhancement method, characterized in that, include: Step 1: Collect the user's fingerprint data using a fingerprint sensor and generate a multidimensional operation dataset; Step 2: Based on the multidimensional operation dataset, extract at least one additional dimension of information; Step 3: Combine and verify the fingerprint data with additional dimension information according to the pre-stored verification rules; Step 4: If the combined verification result matches the pre-stored criteria, then authorize the user to access; otherwise, deny access.

2. The method according to claim 1, characterized in that, The multidimensional operational dataset includes fingerprint image data, coordinate location data, and time series data.

3. The method according to claim 2, characterized in that, The coordinate position data is obtained through two-dimensional coordinate acquisition using a fingerprint sensor.

4. The method according to claim 3, characterized in that, The additional dimension signal includes orientation dimension information, time dimension information, or quantity / area dimension information.

5. The method according to claim 4, characterized in that, The directional dimension information is determined by analyzing changes in coordinate position data to determine the coordinate direction flow.

6. The method according to claim 5, characterized in that, The direction of flow is determined by analyzing changes in coordinate position data. Multiple frames of coordinate data are continuously collected using a fingerprint sensor; Identify the type of data in each frame; The coordinate direction flow is determined based on the type change of adjacent frames.

7. The method according to claim 5, characterized in that, The direction of flow is determined by analyzing changes in coordinate position data. Multiple frames of coordinate data are continuously collected using a fingerprint sensor; The coordinate direction flow is determined based on the time series changes of multi-needle coordinate data.

8. The method according to claim 4, characterized in that, The time dimension information is determined by calculating the touch duration of the fingerprint sensor, including long touches, short touches, or specific time series.

9. The method according to claim 4, characterized in that, The quantity / area dimension information is determined by identifying the effective touch area or combined pattern of the fingerprint sensor.

Images