A method for extracting fingerprint images

Abstract

The present invention discloses a method for obtaining a fingerprint image in original size, and belongs to the technical field of automatic fingerprint recognition. The present invention uses the computing processing power and camera of a mobile computing device to adjust and correct the size, direction and precision of the image by placing a ruler with a standard scale next to the target object, so that the image finally captured can maintain the original size according to the precision required for use. The adjusted original fingerprint image can be sent to a fingerprint or other biometric matching system for direct matching to find the best candidate or the biometric owner of the fingerprint. The fingerprint image is kept in original size, which means that under a given image resolution, according to the image resolution points containing the photographed object, the size value of the photographed object in the image is consistent with the actual size value of the photographed object through the conversion of precision and length. Resolution is usually expressed in "X points / inch".

Description

Technical Field

[0001] The present invention relates to the technical field of automatic fingerprint recognition, and more specifically, to a method for extracting a fingerprint image.

[0002] Keeping the original size of the fingerprint image means that under a given image resolution, the size of the object in the image is consistent with the actual size of the object through the conversion of precision and length according to the image resolution points containing the object being photographed. Resolution is usually expressed in "dots / inch". Background Art

[0003] Fingerprint recognition matching can be divided into two categories: 1:1 matching and 1:N matching. In 1:1 matching, there is only one registered fingerprint on a device, and the fingerprint of the user who attempts to use the device is matched with the registered fingerprint to verify whether the user is legitimate. A common example is using fingerprint verification to unlock a door lock or mobile phone. 1:N matching involves searching a large database containing hundreds of thousands or even hundreds of millions of fingerprint images for one fingerprint to determine the owner of the fingerprint. An important use case of 1:N matching is fingerprint matching, where "non-imprint" fingerprints refer to impressions left accidentally rather than intentionally extracted from an individual. Generally, intentionally extracted fingerprints are called "imprint fingerprints." A common example includes fingerprint matching at the crime scene. The difficulties faced by 1:N matching mainly include the incompleteness and deformation of the fingerprint image, and the change in the ratio when extracting and photographing the fingerprint.

[0004] When the fingerprint database is small or medium-sized, traditional fingerprint recognition technology has been successfully used to solve crimes. The traditional process of using fingerprints to solve crimes usually includes three steps: non-print fingerprint extraction, forensic expert marking features and matching back-end data.

[0005] Although fingerprint matching has been widely used to solve crimes, fingerprint extraction still requires a lot of effort. In many cases, it is usually necessary to place a scale ruler next to the extracted fingerprint, and then use a camera to take a picture of the fingerprint. The image is taken to the fingerprint identification laboratory, and the expert needs to use the tool software in the fingerprint automatic identification system to rescale the image to the original 1:1 scale according to the scale in the image. This requires a lot of manual work and rich work experience. During the entire fingerprint image acquisition process, if the fingerprint is not captured correctly, for example, the camera lens surface is not parallel to the plane where the fingerprint is located, the fingerprint image captured will be deformed, and these deformations are difficult to correct. This deformation will reduce the accuracy of the computer's automatic fingerprint recognition and increase the misidentification rate. In many cases, the fingerprint sample is transferred to a specific fingerprint card by tape, but this process is likely to destroy the original fingerprint pattern or shape, which is also one of the main factors causing deformation. After the sample is collected, the fingerprint sample can be imaged with a camera using the above method, or a flatbed scanner with specific parameters can be used to obtain a 1:1 or original size digital copy of the sample. However, when scanning and entering images later, the scale of each image must be set manually, and the accuracy of the scale parameter setting depends on the technical ability and sense of responsibility of the operator. This makes it very likely that errors will be increased during the critical step of scanning and entering the database.

[0006] The current fingerprint extraction methods require many operation links, each of which may cause distortion and deformation of the fingerprint image, thereby reducing the recognition accuracy of the background fingerprint matching algorithm. With the improvement of work convenience of mobile computing devices such as mobile phones and tablets, they are becoming more and more popular among users in industries such as public security. Automatic acquisition of original fingerprint images on mobile computing devices can lower the technical requirements for operators, reduce the operation links of fingerprint extraction, and thus reduce the negative impact on the quality of fingerprint images, which is an urgent need for fingerprint workers. Summary of the invention

[0007] In view of the problems existing in the prior art, the object of the present invention is to provide a method for obtaining a fingerprint image while maintaining its original size.

[0008] 1. Technical solution

[0009] To solve the above problems, the present invention adopts the following technical solutions.

[0010] A method for extracting an original large fingerprint image comprises the following steps:

[0011] S1. Use the camera of the mobile computing device to calibrate the target fingerprint before taking a photo: by adjusting the distance between the camera and the fingerprint being photographed and zooming in or out on the imaging screen, make the observed length of a certain section of the physical scale and the corresponding millimeter value consistent with the preset length and millimeter number of the interception frame or reference identification line.

[0012] S2 maintains the state after S1 calibration, adjusts the shooting angle to a suitable position so that the target fingerprint is located in a suitable position within the capture frame, and takes a photo.

[0013] S3 captures the image and converts it according to the required image resolution.

[0014] S4 saves the image containing the original fingerprint and transmits it to other application systems in the background for subsequent application.

[0015] Furthermore, in S1, the preset rectangular interception frame or reference identification line is used to identify the length, and is set based on the screen resolution accuracy, and its position and length and width do not change with the enlargement or reduction of the imaging picture.

[0016] Furthermore, the S1 specifically includes the following steps:

[0017] S11, adjusting the position and angle of the camera so that the camera shooting angle is located above the surface to be shot, so that the imaging picture includes a complete target image and part of the real object scale. Auxiliary equipment such as light source can be used during the shooting process to improve the imaging quality.

[0018] S12, focus to make the image clear. The observed physical scale is parallel to or overlaps with the interception frame or reference identification line. By zooming in and out of the image, the length and millimeter value of a certain section of the physical scale are consistent with the preset length and millimeter value of the interception frame or reference identification line.

[0019] Furthermore, in S2, while keeping the shooting distance and the magnification factor unchanged, the camera is rotated in the same horizontal plane so that the photographed fingerprint is located at a suitable position in the capture frame.

[0020] Furthermore, in S3, the captured photo is captured according to the position of the capture frame, and the image in the capture frame is retained. The captured image is converted into a resolution according to the use requirements so that it meets the accuracy requirements of the original image. Furthermore, in S4, the image can be saved to the internal storage space of the mobile computing device, the storage space of an external device, or the storage space based on the network.

[0021] Furthermore, in S4, the image may be saved to an internal storage space of the mobile computing device or a storage space of an external device or a network-based storage space.

[0022] 2. Beneficial effects

[0023] Compared with the prior art, the advantages of the present invention are:

[0024] The present invention provides a method for directly generating a fingerprint image of the original size that meets the requirements when taking a photo using a mobile computing device, which can be directly used in a fingerprint matching application system, thereby reducing work links and improving work efficiency.

[0025] The present invention uses the camera in the mobile computing device to image the target fingerprint. By adjusting the shooting distance and zooming the image, the image can be quickly calibrated and adjusted to the original scale, so as to directly shoot and obtain the original fingerprint image. The operator can use a handheld device or other fixed objects to adjust and maintain the shooting distance and stabilize the shooting device to avoid deformation and disproportion. After calibration, maintaining the same shooting distance and zoom factor can more quickly shoot the original fingerprint image. After the original fingerprint image is uploaded to the back-end fingerprint recognition system, it can be directly used for comparison and accurately matched.

[0026] The ease of use, accuracy and efficiency of the present invention:

[0027] The present invention is an improvement and enhancement over the traditional method. The present invention uses the computing power of the computing device itself and drives the camera device, so that the zoom, calibration and shooting functions are integrated together, and the operator can quickly complete the calibration and shooting process. While maintaining the same distance and zoom factor, the original fingerprint can be shot more quickly and continuously. The captured image can be directly transmitted to the back-end fingerprint matching system for calculation and matching without manual adjustment and calibration again, so as to find the best candidate to match it or directly find the owner of the fingerprint. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flow chart of the method steps of the present invention;

[0029] Figure 2 This is a specific operation flow chart for taking fingerprint photos in the present invention;

[0030] Figure 3 It is a schematic diagram of the present invention when performing scaling factor calibration;

[0031] Figure 4 This is a schematic diagram of imaging when photographing a scale in the present invention. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the specification of the present invention. The described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work belong to the protection scope of the present invention.

[0033] The embodiment is divided into four parts:

[0034] 1. Set the capture box and reference identification line segment.

[0035] 2. Adjust the shooting distance and zoom ratio, align the length of the actual scale with the interception frame or reference identification line segment, and take a photo.

[0036] 3. Capture the image, convert the image resolution as required, and save the original fingerprint image.

[0037] 4. Send to the fingerprint matching application system for fingerprint matching.

[0038] Specifically:

[0039] Given a standard example, according to the GA standard requirements of the Ministry of Public Security, the size of the on-site fingerprint image is 512×512dpi and the resolution is 500dpi / inch. Based on this image accuracy requirement, in the camera imaging picture on the display screen of the mobile computing device, the side length of the preset square interception frame is twice the required accuracy, that is, the side length is 1000 points. The required accuracy of 500dpi is 500 dots / inch, and the length of 512 points converted to metric is 26 mm. Based on this, the total scale length of the interception frame is set to 26 mm. Similarly, based on this image accuracy, the total scale length of the calibration reference segment can be set to 15 mm, and the calibration reference segment on the screen is 577 points in length.

[0040] Adjust the shooting distance and zoom, hold the mobile computing device or place it on an object to control the shooting distance. Focus and make the image clear, then use the zoom function to make the observed physical scale length consistent with the scale length marked by the interception frame or the calibration reference line segment, specifically, make the 26 mm length on the physical scale the same as the length of one side of the interception frame, or make the 15 mm length on the physical scale the same as the 15 mm standard reference line segment. Take a photo.

[0041] The photographed image is cropped according to the range shown in the capture frame, and the image within the capture frame is retained. The captured image is converted to a resolution of 500 dpi, thereby obtaining a fingerprint image that retains its original size. The image is saved.

[0042] The saved image can be transmitted to the backend application system via the network or storage medium for direct fingerprint comparison.

[0043] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person familiar with the technical field can make equivalent replacement or changes according to the technical solution and its improved conception within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for extracting an original fingerprint image, characterized in that: S1. Calibrate the target fingerprint before taking a photo using the camera of the mobile computing device: by adjusting the distance between the camera and the fingerprint being photographed and zooming in or out on the image, make the length of a certain section of the observed physical scale and the corresponding millimeter value consistent with the preset length and millimeter value of the interception frame or reference identification line; S2, keep the state after S1 calibration, adjust the shooting angle to make the target fingerprint be located at a suitable position in the interception frame, and take a photo; S3, capturing the image and converting it according to the image resolution required for use; S4, saving the image of the fingerprint that is consistent with the actual size of the photographed fingerprint, and transmitting it to other background application systems for subsequent application; In S1, the preset rectangular interception frame or reference identification line is used to identify the length, and is set based on the screen resolution accuracy. Its position and length and width do not change with the enlargement or reduction of the imaging picture.

2. The method for extracting an original fingerprint image according to claim 1, characterized in that: The S1 specifically includes the following steps: S11, adjusting the position and angle of the camera so that the camera shooting angle is located above the surface to be shot, so that the imaging picture contains a complete target image and part of the real object scale; using auxiliary equipment such as light source during the shooting process to improve the imaging quality; S12. Focus to make the image clear; the observed physical scale is parallel to or overlaps with the interception frame or reference identification line, and by zooming in and out of the image, the length and millimeter value of a certain section of the physical scale are made consistent with the preset length and millimeter value of the interception frame or reference identification line.

3. The method for extracting an original fingerprint image according to claim 1, characterized in that: In S2, while keeping the shooting distance and the magnification factor unchanged, the camera is rotated in the same horizontal plane so that the photographed fingerprint is located at a suitable position in the capture frame.

4. The method for extracting an original fingerprint image according to claim 1, characterized in that: In S3, the captured photo is captured according to the position of the capture frame, and the image within the capture frame is retained; the captured image is converted into a resolution according to the use requirements, so that it meets the accuracy requirements of the image whose required size value is consistent with the actual size value of the photographed object.

5. The method for extracting an original fingerprint image according to claim 1, characterized in that: In S4, the image is saved to the internal storage space of the mobile computing device or the storage space of an external device or a network-based storage space.

Images