Self-adaptive gain adjustment fingerprint image enhancement method

Abstract

The invention relates to a self-adaptive gain adjustment fingerprint image enhancement method, which comprises the following steps of: firstly, performing frequency domain decomposition on a fingerprint image to obtain a low-frequency component and a high-frequency component; then calculating a local variance and a local standard deviation of the high-frequency component, and adaptively determining a gain adjustment coefficient according to the standard deviation; performing adaptive enhancement on the high-frequency component by using a gain adjustment coefficient; and finally, recombining the enhanced high-frequency component and the enhanced low-frequency component to obtain an enhanced fingerprint image. Through frequency domain decomposition and adaptive gain adjustment, the definition and contrast of the fingerprint image can be effectively improved, detail features of the fingerprint are highlighted, a good foundation is laid for subsequent fingerprint feature extraction and comparison, and the accuracy of fingerprint identification is improved.

Description

Technical Field

[0001] This invention relates to the field of fingerprint recognition, and in particular to a fingerprint image enhancement method with adaptive gain adjustment. Background Technology

[0002] Fingerprint recognition technology, with its advantages of uniqueness, stability, and non-invasiveness, has been widely used in personal identification and security. However, the quality of fingerprint images is affected by factors such as skin condition and acquisition equipment, often resulting in insufficient clarity and low contrast, which poses difficulties for fingerprint feature extraction and comparison. To improve the accuracy of fingerprint recognition, fingerprint image enhancement processing is necessary.

[0003] Existing fingerprint image enhancement methods can be mainly categorized as follows: The first category is spatial domain-based methods, such as histogram equalization and Retinex algorithm. These methods improve contrast by adjusting the grayscale distribution of the image, but their enhancement effect on image details is limited. The second category is frequency domain-based methods, such as Fourier transform and wavelet transform. These methods enhance the image by processing the frequency domain components, but may introduce artifacts such as ringing during the processing. The third category is directional filtering methods, such as Gabor filtering and directional filtering. These methods enhance the image by utilizing the directional information of fingerprint ridges, but their processing effect on areas such as breaks and intersections of fingerprint ridges is poor.

[0004] In summary, existing fingerprint image enhancement methods, while improving image quality, often suffer from drawbacks such as insufficient detail enhancement, artifact introduction, and weak adaptability. This is mainly because existing methods do not effectively utilize the frequency domain characteristics and local adaptive properties of fingerprint images, failing to effectively suppress noise while enhancing fingerprint details. Therefore, it is necessary to research a new fingerprint image enhancement method that can effectively improve the clarity of fingerprint details while adaptively suppressing noise and artifacts, thereby achieving a more ideal enhancement effect. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes an adaptive gain adjustment fingerprint image enhancement method to solve the issues of poor fingerprint image enhancement effect and weak adaptability in existing technologies. This method adaptively adjusts the gain coefficient of local regions, improving fingerprint detail contrast while effectively suppressing noise amplification and artifact generation, thereby obtaining a clearer and more accurate fingerprint image.

[0006] An adaptive gain adjustment fingerprint image enhancement method includes the following steps: Step 1: Input a grayscale image of your fingerprint; Step 2: Perform frequency domain decomposition on the input fingerprint image to obtain the low-frequency part representing the background and the high-frequency part representing the details; Step 3: Enhance the high-frequency component to obtain the enhanced high-frequency component; Step 4: Recombine the enhanced high-frequency part with the low-frequency part to obtain the final enhanced fingerprint image.

[0007] Furthermore, step 2 specifically includes: (1) Taking each pixel (i, j) in the input fingerprint grayscale image I as the center, select a window W of size (2n+1)×(2n+1), calculate the local mean Low(i, j) within the window, and obtain the low-frequency component. The calculation formula is as follows: Where I(i′, j′) represents the pixel value at coordinates (i′, j′) within window W, and n is a non-negative integer representing the parameter of window size; (2) The high-frequency component High(i,j) of the fingerprint image is obtained by subtracting the low-frequency component Low from the original fingerprint grayscale image I. The calculation formula is as follows: .

[0008] Furthermore, step 3 includes the following sub-steps: (1) Taking each pixel (i, j) in the input fingerprint grayscale image I as the center, select a window W of size (2n+1)×(2n+1) and calculate the local variance Var(i, j) within the window: Where I(i', j') represents the pixel value at coordinates (i', j') within window W, and Low(i, j) represents the local mean within window W; (2) Take the square root of the local variance to obtain the local standard deviation Std(i, j): (3) Calculate the adaptive gain coefficient. The gain coefficient G(i, j) for each pixel is: Where D is a preset constant used to control the range of values ​​for the gain coefficient; (4) Adaptive gain adjustment is performed on the high-frequency components. The adaptive gain coefficient is multiplied by the high-frequency components to obtain the enhanced high-frequency components HighEnhanced(i, j). The calculation formula is as follows: Here, High(i, j) represents the high-frequency components of the original image. When the enhanced high-frequency components exceed 255, they are limited to 255; when the enhanced high-frequency components are less than 0, they are limited to 0; otherwise, the enhanced high-frequency component values ​​are used directly.

[0009] Furthermore, step 4 specifically includes: The enhanced high-frequency component is added to the low-frequency component to obtain the final enhanced fingerprint image Enhanced(i,j), calculated using the following formula: .

[0010] The beneficial effects of this invention are as follows: By performing frequency domain decomposition on fingerprint images and adaptively adjusting the gain of the high-frequency components, this invention effectively enhances the details and texture information of fingerprint images, resulting in clearer and sharper images with significant enhancement effects. This invention introduces the calculation of local variance and local standard deviation, enabling adaptive adjustment of the gain coefficient based on the local features of the fingerprint image. This allows the enhancement effect to adapt to fingerprint images of different qualities and types, exhibiting strong adaptability and robustness. The use of frequency domain decomposition and local variance calculation methods avoids the convolution operation in traditional spatial domain enhancement methods, greatly reducing computational complexity and improving the algorithm's computational efficiency, enabling real-time or near-real-time fingerprint image enhancement processing. The fingerprint image enhancement method proposed in this invention can not only be used in the field of fingerprint recognition but can also be extended to other biometric recognition, medical image processing, remote sensing image enhancement, and other fields, showing broad application prospects. Attached Figure Description

[0011] Figure 1 This is a flowchart of an adaptive gain adjustment fingerprint image enhancement method according to the present invention.

[0012] Figure 2 This is an input fingerprint image.

[0013] Figure 3 The fingerprint image enhanced using the present invention. Detailed Implementation

[0014] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0015] Example 1.

[0016] like Figure 1 As shown, this embodiment provides a fingerprint image enhancement method with adaptive gain adjustment, including the following steps: Step 1: Input a grayscale image of your fingerprint; Step 2: Perform frequency domain decomposition on the input fingerprint image to obtain the low-frequency component representing the background and the high-frequency component representing the details. Specifically, for each pixel (i, j) in the input fingerprint grayscale image I, select a window W of size (2n+1)×(2n+1), calculate the local mean Low(i, j) within the window, and obtain the low-frequency component. The calculation formula is as follows: Where I(i′, j′) represents the pixel value at coordinates (i′, j′) within window W, and n is a non-negative integer representing the parameter of window size.

[0017] The high-frequency component High(i,j) of the fingerprint image is obtained by subtracting the low-frequency component Low from the original fingerprint grayscale image I. The calculation formula is as follows: .

[0018] Step 3: Enhance the high-frequency components to obtain the enhanced high-frequency components; specifically, for each pixel (i, j) in the input fingerprint grayscale image I, select a window W of size (2n+1)×(2n+1), and calculate the local variance Var(i, j) within this window: Where I(i', j') represents the pixel value at coordinates (i', j') within window W, and Low(i, j) represents the local mean within window W.

[0019] Taking the square root of the local variance yields the local standard deviation Std(i, j): .

[0020] Calculate the adaptive gain coefficients, where the gain coefficient G(i, j) for each pixel is: Where D is a preset constant used to control the range of values ​​for the gain coefficient.

[0021] Adaptive gain adjustment is performed on the high-frequency components. The adaptive gain coefficient is multiplied by the high-frequency components to obtain the enhanced high-frequency components HighEnhanced(i, j). The calculation formula is as follows: Here, High(i, j) represents the high-frequency components of the original image. When the enhanced high-frequency components exceed 255, they are limited to 255; when the enhanced high-frequency components are less than 0, they are limited to 0; otherwise, the enhanced high-frequency component values ​​are used directly.

[0022] Step 4: Reconstruct the enhanced high-frequency components with the low-frequency components to obtain the final enhanced fingerprint image; specifically, add the enhanced high-frequency components to the low-frequency components to obtain the final enhanced fingerprint image Enhanced(i,j), calculated using the following formula: .

[0023] The fingerprint image enhancement method with adaptive gain adjustment provided in this embodiment improves the contrast of fingerprint details while effectively suppressing noise amplification and artifact generation, thereby obtaining a clearer and more accurate fingerprint image by adaptively adjusting the gain coefficient of local regions.

[0024] It should be noted that the value of n in the above embodiments can be set according to actual needs, for example, it can be set to 1, 2, 3, etc. In addition, the constant D in the above embodiments can also be set according to actual needs, for example, it can be set to 10, 20, 30, etc.

[0025] Those skilled in the art will understand that the steps in the above embodiments can be adjusted and optimized according to actual needs. For example, the execution order of each step can be adjusted, or some steps can be combined for execution, as long as the technical solution of the present invention can be achieved.

[0026] The above description is only a preferred embodiment of the present invention, but the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A fingerprint image enhancement method with adaptive gain adjustment, characterized in that, Includes the following steps: Step 1: Input a grayscale image of your fingerprint; Step 2: Perform frequency domain decomposition on the input fingerprint image to obtain the low-frequency part representing the background and the high-frequency part representing the details; Step 3: Enhance the high-frequency component to obtain the enhanced high-frequency component; Step 4: Recombine the enhanced high-frequency part with the low-frequency part to obtain the final enhanced fingerprint image.

2. The method according to claim 1, characterized in that, Step 2 includes the following sub-steps: (1) Taking each pixel (i, j) in the input fingerprint grayscale image I as the center, select a window W of size (2n+1)×(2n+1), calculate the local mean Low(i, j) within the window, and obtain the low-frequency component. The calculation formula is as follows: Where I(i′, j′) represents the pixel value at coordinates (i′, j′) within window W, and n is a non-negative integer representing the parameter of window size; (2) The high-frequency component High(i, j) of the fingerprint image is obtained by subtracting the low-frequency component Low from the original fingerprint grayscale image I. The calculation formula is as follows: 。 3. The method according to claim 1, characterized in that, Step 3 includes the following sub-steps: (1) Taking each pixel (i, j) in the input fingerprint grayscale image I as the center, select a window W of size (2n+1)×(2n+1) and calculate the local variance Var(i, j) within the window: Where I(i', j') represents the pixel value at coordinates (i', j') within window W, and Low(i, j) represents the local mean within window W; (2) Take the square root of the local variance to obtain the local standard deviation Std(i, j): (3) Calculate the adaptive gain coefficient. The gain coefficient G(i, j) for each pixel is: Where D is a preset constant used to control the range of values ​​for the gain coefficient; (4) Adaptive gain adjustment is performed on the high-frequency components. The adaptive gain coefficient is multiplied by the high-frequency components to obtain the enhanced high-frequency components HighEnhanced(i, j): Here, High(i, j) represents the high-frequency components of the original image. When the enhanced high-frequency components exceed 255, they are limited to 255; when the enhanced high-frequency components are less than 0, they are limited to 0; otherwise, the enhanced high-frequency component values ​​are used directly.

4. The method according to claim 1, characterized in that, Step 4 includes the following steps: The enhanced high-frequency component is added to the low-frequency component to obtain the final enhanced fingerprint image Enhanced(i,j), calculated using the following formula:

Images