A fingerprint recognition method and system based on optical recognition sensing technology

Abstract

The present invention discloses a fingerprint recognition method and system based on optical recognition sensing technology, which relates to the field of fingerprint recognition technology, including: using the total reflection principle of light to collect fingerprint surface texture images; converting the collected fingerprint surface texture images into fingerprint surface texture grayscale images; determining the main direction and frequency of fingerprint lines, and separating the adhering fingerprint lines; using image filtering algorithms to smooth out the stray point information in the fingerprint image; using morphological expansion and corrosion operations to further highlight the undulating characteristics of the fingerprint; extracting the morphological characteristics of the fingerprint and the local characteristics of the lines from the preprocessed fingerprint image; and outputting the fingerprint corresponding user information in the database that is higher than the preset similarity value. By setting an image acquisition module, an image processing module, a feature extraction module, and a similarity comparison module, the interference of environmental factors is reduced, the quality of the fingerprint image is ensured, the fingerprint recognition time is shortened, and the real-time performance of the system is improved.

Description

Technical Field

[0001] The present invention relates to the field of fingerprint recognition technology, and in particular to a fingerprint recognition method and system based on optical recognition sensing technology. Background Art

[0002] Fingerprint collection and recognition technology has a wide range of applications, especially in situations where high-precision identity authentication is required. In the logistics industry, fingerprint recognition technology can be used in warehouse management, express delivery, and cargo tracking. Through fingerprint recognition, companies can ensure that only authorized personnel can enter specific areas, effectively preventing the loss and theft of items. At the same time, in the express delivery process, fingerprint recognition technology can be used for recipient identity authentication to ensure that the express is safely and accurately delivered to the designated person. With the popularity of smart phones, fingerprint recognition has become a standard feature for unlocking mobile phones. Users only need to press their fingers lightly on the fingerprint recognition area to achieve one-click unlocking, which greatly improves the security and convenience of the mobile phone.

[0003] Traditional fingerprint collection technology is easily affected by multiple factors such as environment, equipment, and finger condition, resulting in unstable collection quality. Fingerprint images are often affected by multiple factors such as noise and light changes during the collection process, resulting in reduced image quality. Fingerprint recognition technology involves the user's biometric information, so privacy protection issues have attracted much attention. When applying fingerprint collection and recognition technology, it is necessary to ensure that the user's fingerprint information is secure and not leaked or abused. Summary of the invention

[0004] In order to solve the above technical problems, a fingerprint recognition method based on optical recognition sensing technology is provided. This technical solution solves the problems raised in the above background technology.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A fingerprint recognition method based on optical recognition sensing technology, comprising:

[0007] The acquisition device illuminates the finger through the built-in light source, and based on the different reflectivity of the ridges and valleys of the fingerprint, uses the principle of total reflection of light to collect the fingerprint surface texture image;

[0008] The collected fingerprint surface texture image is converted into a fingerprint surface texture grayscale image, and the fingerprint is separated from the background to remove the background area;

[0009] By calculating the gradient of the grayscale image of the fingerprint surface texture, the main direction and frequency of the fingerprint lines are determined, and the adhering fingerprint lines are separated;

[0010] Use image filtering algorithm to smooth out the stray point information in the fingerprint image and improve the ridge information of the fingerprint image;

[0011] Morphological dilation and erosion operations are used to further highlight the undulating features of fingerprints;

[0012] Extracting fingerprint morphological features and local features of ridges from the preprocessed fingerprint image, wherein the fingerprint morphological features include undulation, bifurcation and intersection of the fingerprint, and the local features of ridges include direction, length and spacing of the ridges;

[0013] The extracted fingerprint features are compared with the fingerprint features stored in the database, and the similarity between the input fingerprint and the fingerprint in the database is calculated using a similarity measurement algorithm, and the user information corresponding to the fingerprint in the database that is higher than the preset similarity value is output.

[0014] Preferably, the step of converting the collected fingerprint surface texture image into a fingerprint surface texture grayscale image and separating the fingerprint from the background, and removing the background area specifically includes:

[0015] Obtain the weighted coefficients of human eye sensitivity to red, green and blue;

[0016] Substitute the red, green and blue color components of each pixel in the fingerprint surface texture image into the weighted average formula to obtain the grayscale value of each pixel in the fingerprint surface texture grayscale image;

[0017] Based on the gray value distribution of the gray image of the fingerprint surface texture, a gray value threshold is preset;

[0018] Determine whether the grayscale value of each pixel in the grayscale image of the fingerprint surface texture is higher than the grayscale value threshold. If so, mark the pixel as the fingerprint area; if not, mark the pixel as the background area;

[0019] Remove the background area in the grayscale image of the fingerprint surface texture;

[0020] The weighted average formula is: ,

[0021] In the formula, is the gray value of the pixel in the i-th row and j-th column in the gray image of the fingerprint surface texture, are the weighted coefficients of human eye sensitivity to red, green and blue, respectively. They are respectively the red, green and blue color components of the pixel at the i-th row and j-th column in the fingerprint surface texture image.

[0022] Preferably, the method of determining the main direction and frequency of the fingerprint lines by performing gradient calculation on the grayscale image of the fingerprint surface texture and separating the adhering fingerprint lines specifically includes:

[0023] The horizontal gradient operator is used to perform convolution operation on the grayscale image of the fingerprint surface texture, and the result of the convolution operation is output as the gradient of the grayscale image of the fingerprint surface texture in the x direction;

[0024] The vertical gradient operator is used to perform convolution operation on the grayscale image of the fingerprint surface texture. The result of the convolution operation is output as the gradient of the grayscale image of the fingerprint surface texture in the y direction.

[0025] By calculating the square root of the sum of the squares of the gradients in the x-direction and the y-direction, the gradient of each pixel in the grayscale image of the fingerprint surface texture is obtained;

[0026] By calculating the inverse tangent value of the y-direction gradient and the x-direction gradient, the gradient direction of each pixel in the grayscale image of the fingerprint surface texture is obtained;

[0027] By sliding a 3×3 window on the fingerprint image, the periodic change of the gradient of each pixel in the window is calculated and output as the frequency;

[0028] Perform statistical analysis on the directions of all pixels, and output the direction with the highest frequency as the main direction of the fingerprint lines;

[0029] Based on the main direction of the fingerprint lines, the fingerprint lines in the main direction are enhanced and the adhesion phenomenon in the non-main direction is reduced;

[0030] The convolution operation is: ,

[0031] In the formula, is the gradient of the pixel at the i-th row and j-th column in the grayscale image of the fingerprint surface texture. When n is x, it is the gradient in the x direction. When n is y, it is the gradient in the y direction. is the gradient operator. When n is x, the gradient operator is , when n is y, the gradient operator is .

[0032] Preferably, the method of smoothing out the stray point information in the fingerprint image and improving the ridge information of the fingerprint image by using an image filtering algorithm specifically includes:

[0033] By calculating the average grayscale value of pixels around each pixel in the grayscale image of the fingerprint surface texture, the noise is reduced and the stray point information is filtered out;

[0034] Based on the main direction of the fingerprint lines, the directional parameter value of the filter response is preset;

[0035] Based on the spatial frequency of the filter, the frequency parameter value of the filter is preset;

[0036] Based on the sensitivity to noise and the accuracy requirements of the features, the filter window size parameter value is preset;

[0037] Preset the phase shift parameter value to adjust the phase response of the filter;

[0038] Get the filter kernel function formula;

[0039] The constructed filter is applied to the grayscale image of the fingerprint surface texture. The filter kernel function slides on the image and calculates the convolution result at each position.

[0040] Output the filtered image;

[0041] The filter kernel function formula is: ,

[0042] In the formula, is the filter kernel function, is the directional parameter value of the filter response, is the frequency parameter value of the filter, is the filter window size parameter value, is the phase shift parameter value of the filter.

[0043] Preferably, the use of morphological dilation and erosion operations to further highlight the undulating features of the fingerprint specifically includes:

[0044] Select structural elements whose size and shape match the ridges and valleys characteristic of the fingerprint;

[0045] The defined structural element is used to dilate the grayscale image of the fingerprint surface texture, so that the part of the image with brightness higher than the preset brightness threshold expands outward to highlight the convex features of the fingerprint;

[0046] Check the expanded image to ensure that the ridge line is expanded and avoid image distortion caused by over-expansion;

[0047] Applying an erosion operation on the dilated image causes the portion of the image with brightness higher than a preset brightness threshold to shrink inward, thus highlighting the concave features of the fingerprint;

[0048] Adjust the size and shape of the structural element based on the eroded image effect;

[0049] Check the image after etching to ensure that the valley line is deepened and avoid excessive etching that may cause loss of fingerprint features.

[0050] Dilation and erosion operations were applied alternately to highlight the undulating features of the fingerprint. After each operation, the image effect was checked and the structural element parameters were adjusted.

[0051] Preferably, the step of comparing the extracted fingerprint features with the fingerprint features stored in the database, using a similarity measurement algorithm to calculate the similarity between the input fingerprint and the fingerprint in the database, and outputting the user information corresponding to the fingerprint in the database that is higher than a preset similarity value specifically includes:

[0052] Loading stored fingerprint features and their corresponding user information from the fingerprint database;

[0053] Convert the extracted fingerprint features and the stored fingerprint features into non-zero feature vector representations;

[0054] Based on the cosine similarity algorithm, the ratio of the product of the Euclidean norm of the non-zero feature vector converted from the extracted fingerprint feature and the stored fingerprint feature to the dot product between the vectors is calculated and output as the fingerprint similarity;

[0055] Determine whether the similarity of the fingerprint is higher than the preset similarity threshold. If so, output the user information corresponding to the fingerprint in the database; if not, do not output it.

[0056] Furthermore, a fingerprint recognition system based on optical recognition sensing technology is proposed, which is used to implement the fingerprint recognition method based on optical recognition sensing technology as described above, including:

[0057] An image acquisition module, wherein the image acquisition module is used to collect fingerprint surface texture images by illuminating the fingerprint through a built-in light source and utilizing the principle of total reflection of light based on the different reflectivity of the fingerprint ridges and valleys to light;

[0058] An image processing module is used to convert the collected fingerprint surface texture image into a fingerprint surface texture grayscale image, separate the fingerprint from the background, remove the background area, determine the main direction and frequency of the fingerprint lines by performing gradient calculation on the fingerprint surface texture grayscale image, separate the adhering fingerprint lines, smooth out the stray point information in the fingerprint image by using an image filtering algorithm, and improve the ridge information of the fingerprint image, and use morphological expansion and corrosion operations to further highlight the undulating characteristics of the fingerprint;

[0059] A feature extraction module, which is used to extract the morphological features of the fingerprint and the local features of the ridges from the preprocessed fingerprint image;

[0060] The similarity comparison module is used to compare the extracted fingerprint features with the fingerprint features stored in the database, use the similarity measurement algorithm to calculate the similarity between the input fingerprint and the fingerprint in the database, and output the user information corresponding to the fingerprint in the database that is higher than the preset similarity value.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] Based on the principle of total reflection, it can efficiently and accurately collect the texture image of the fingerprint surface, improve the clarity of fingerprint collection, enhance the reliability of fingerprint recognition, effectively separate the adhered fingerprint lines, avoid recognition errors caused by line adhesion, highlight the undulating characteristics of the fingerprint, enhance the key information in the fingerprint image, improve the accuracy of fingerprint recognition, and can quickly and accurately find the database record that matches the input fingerprint, thereby improving the security of fingerprint recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of a fingerprint recognition method based on optical recognition sensing technology of the present invention;

[0064] Figure 2 A flow chart of the method for converting grayscale images and removing background areas of the present invention;

[0065] Figure 3 This is a flow chart of the method for determining the main direction and frequency of fingerprint lines and separating the adhering fingerprint lines of the present invention;

[0066] Figure 4 This is a flow chart of a method for smoothing out stray point information in a fingerprint image using an image filtering algorithm according to the present invention;

[0067] Figure 5 A flow chart of a method for further highlighting the undulating characteristics of fingerprints by using morphological dilation and erosion operations according to the present invention;

[0068] Figure 6 This is a flow chart of a method for calculating the similarity between an input fingerprint and a fingerprint in a database using a similarity measurement algorithm according to the present invention. DETAILED DESCRIPTION

[0069] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0070] Reference Figure 1 As shown, a fingerprint recognition method based on optical recognition sensing technology includes:

[0071] The acquisition device illuminates the finger through the built-in light source, and based on the different reflectivity of the ridges and valleys of the fingerprint, uses the principle of total reflection of light to collect the fingerprint surface texture image;

[0072] The collected fingerprint surface texture image is converted into a fingerprint surface texture grayscale image, and the fingerprint is separated from the background to remove the background area;

[0073] By calculating the gradient of the grayscale image of the fingerprint surface texture, the main direction and frequency of the fingerprint lines are determined, and the adhering fingerprint lines are separated;

[0074] Use image filtering algorithm to smooth out the stray point information in the fingerprint image and improve the ridge information of the fingerprint image;

[0075] Morphological dilation and erosion operations are used to further highlight the undulating features of fingerprints;

[0076] Extracting fingerprint morphological features and local features of ridges from the preprocessed fingerprint image, wherein the fingerprint morphological features include undulation, bifurcation and intersection of the fingerprint, and the local features of ridges include direction, length and spacing of the ridges;

[0077] The extracted fingerprint features are compared with the fingerprint features stored in the database, and the similarity between the input fingerprint and the fingerprint in the database is calculated using a similarity measurement algorithm, and the user information corresponding to the fingerprint in the database that is higher than the preset similarity value is output.

[0078] Reference Figure 2 As shown, the collected fingerprint surface texture image is converted into a fingerprint surface texture grayscale image, and the fingerprint is separated from the background. Removing the background area specifically includes:

[0079] Obtain the weighted coefficients of human eye sensitivity to red, green and blue;

[0080] Substitute the red, green and blue color components of each pixel in the fingerprint surface texture image into the weighted average formula to obtain the grayscale value of each pixel in the fingerprint surface texture grayscale image;

[0081] Based on the gray value distribution of the gray image of the fingerprint surface texture, a gray value threshold is preset;

[0082] Determine whether the grayscale value of each pixel in the grayscale image of the fingerprint surface texture is higher than the grayscale value threshold. If so, mark the pixel as the fingerprint area; if not, mark the pixel as the background area;

[0083] Remove the background area in the grayscale image of the fingerprint surface texture;

[0084] The weighted average formula is: ,

[0085] In the formula, is the gray value of the pixel in the i-th row and j-th column in the gray image of the fingerprint surface texture, are the weighted coefficients of human eye sensitivity to red, green and blue, respectively. They are respectively the red, green and blue color components of the pixel at the i-th row and j-th column in the fingerprint surface texture image.

[0086] During the collection process, fingerprint images are often affected by multiple factors such as noise and lighting changes, resulting in a decrease in image quality. By converting the collected fingerprint surface texture image into a grayscale image and separating it from the background, the interference of the background area can be removed and the impact of noise and lighting changes on the image can be reduced.

[0087] Reference Figure 3 As shown, by calculating the gradient of the grayscale image of the fingerprint surface texture, the main direction and frequency of the fingerprint lines are determined, and the separation of the adhering fingerprint lines specifically includes:

[0088] The horizontal gradient operator is used to perform convolution operation on the grayscale image of the fingerprint surface texture, and the result of the convolution operation is output as the gradient of the grayscale image of the fingerprint surface texture in the x direction;

[0089] The vertical gradient operator is used to perform convolution operation on the grayscale image of the fingerprint surface texture. The result of the convolution operation is output as the gradient of the grayscale image of the fingerprint surface texture in the y direction.

[0090] By calculating the square root of the sum of the squares of the gradients in the x-direction and the y-direction, the gradient of each pixel in the grayscale image of the fingerprint surface texture is obtained;

[0091] By calculating the inverse tangent value of the y-direction gradient and the x-direction gradient, the gradient direction of each pixel in the grayscale image of the fingerprint surface texture is obtained;

[0092] By sliding a 3×3 window on the fingerprint image, the periodic change of the gradient of each pixel in the window is calculated and output as the frequency;

[0093] Perform statistical analysis on the directions of all pixels, and output the direction with the highest frequency as the main direction of the fingerprint lines;

[0094] Based on the main direction of the fingerprint lines, the fingerprint lines in the main direction are enhanced and the adhesion phenomenon in the non-main direction is reduced;

[0095] The convolution operation is: ,

[0096] In the formula, is the gradient of the pixel at the i-th row and j-th column in the grayscale image of the fingerprint surface texture. When n is x, it is the gradient in the x direction. When n is y, it is the gradient in the y direction. is the gradient operator. When n is x, the gradient operator is , when n is y, the gradient operator is .

[0097] When performing convolution operations, you need to pay attention to processing image boundaries. Usually, you can use filling (such as zero filling, mirror filling, etc.) or cropping boundary pixels to handle it. Gradient calculation may involve a large number of convolution operations, so you need to pay attention to computational efficiency when processing large-scale fingerprint images.

[0098] Reference Figure 4 As shown in FIG. 1 , the image filtering algorithm is used to smooth out the stray point information in the fingerprint image and improve the ridge information of the fingerprint image. Specifically, the following steps are used:

[0099] By calculating the average grayscale value of pixels around each pixel in the grayscale image of the fingerprint surface texture, the noise is reduced and the stray point information is filtered out;

[0100] Based on the main direction of the fingerprint lines, the directional parameter value of the filter response is preset;

[0101] Based on the spatial frequency of the filter, the frequency parameter value of the filter is preset;

[0102] Based on the sensitivity to noise and the accuracy requirements of the features, the filter window size parameter value is preset;

[0103] Preset the phase shift parameter value to adjust the phase response of the filter;

[0104] Get the filter kernel function formula;

[0105] The constructed filter is applied to the grayscale image of the fingerprint surface texture. The filter kernel function slides on the image and calculates the convolution result at each position.

[0106] Output the filtered image;

[0107] The filter kernel function formula is: ,

[0108] In the formula, is the filter kernel function, is the directional parameter value of the filter response, is the frequency parameter value of the filter, is the filter window size parameter value, is the phase shift parameter value of the filter.

[0109] The image is filtered using a filter, which is a bandpass filter with specific direction and frequency characteristics. It can enhance the fingerprint ridge information that matches the filter direction and frequency, while suppressing noise and interference in other directions. By adjusting the parameters of the filter (such as direction, frequency, bandwidth, etc.), effective enhancement of the fingerprint ridges can be achieved.

[0110] Reference Figure 5As shown in the figure, the morphological dilation and erosion operations are used to further highlight the undulating features of the fingerprint, including:

[0111] Select structural elements whose size and shape match the ridges and valleys characteristic of the fingerprint;

[0112] The defined structural element is used to dilate the grayscale image of the fingerprint surface texture, so that the part of the image with brightness higher than the preset brightness threshold expands outward to highlight the convex features of the fingerprint;

[0113] Check the expanded image to ensure that the ridge line is expanded and avoid image distortion caused by over-expansion;

[0114] Applying an erosion operation on the dilated image causes the portion of the image with brightness higher than a preset brightness threshold to shrink inward, thus highlighting the concave features of the fingerprint;

[0115] Adjust the size and shape of the structural element based on the eroded image effect;

[0116] Check the image after etching to ensure that the valley line is deepened and avoid excessive etching that may cause loss of fingerprint features.

[0117] Dilation and erosion operations were applied alternately to highlight the undulating features of the fingerprint. After each operation, the image effect was checked and the structural element parameters were adjusted.

[0118] The size and shape of the structural element have a great influence on the effect of morphological operations. A structural element that is too large may cause excessive smoothing or loss of fingerprint features, while a structural element that is too small may not be able to effectively highlight the fingerprint features. The number of dilation and erosion operations should be controlled according to the actual situation of the fingerprint image. Too many operations may cause image distortion or feature loss, while too few operations may not achieve the expected feature highlighting effect.

[0119] Reference Figure 6 As shown, the extracted fingerprint features are compared with the fingerprint features stored in the database, and the similarity between the input fingerprint and the fingerprint in the database is calculated using a similarity measurement algorithm, and the user information corresponding to the fingerprint in the database that is higher than the preset similarity value is output, including:

[0120] Loading stored fingerprint features and their corresponding user information from the fingerprint database;

[0121] Convert the extracted fingerprint features and the stored fingerprint features into non-zero feature vector representations;

[0122] Based on the cosine similarity algorithm, the ratio of the product of the Euclidean norm of the non-zero feature vector converted from the extracted fingerprint feature and the stored fingerprint feature to the dot product between the vectors is calculated and output as the fingerprint similarity;

[0123] Determine whether the similarity of the fingerprint is higher than the preset similarity threshold. If so, output the user information corresponding to the fingerprint in the database; if not, do not output it.

[0124] The filtered matching results are further processed. If there is only one matching result, the user information corresponding to the result is directly output. If there are multiple matching results, you can choose to output the most similar user information, output all matching results, or perform further verification according to actual needs.

[0125] Further, based on the same inventive concept as the above-mentioned fingerprint recognition method based on optical recognition sensing technology, the present solution also proposes a fingerprint recognition system based on optical recognition sensing technology, including:

[0126] An image acquisition module, wherein the image acquisition module is used to collect fingerprint surface texture images by illuminating the fingerprint through a built-in light source and utilizing the principle of total reflection of light based on the different reflectivity of the fingerprint ridges and valleys to light;

[0127] An image processing module is used to convert the collected fingerprint surface texture image into a fingerprint surface texture grayscale image, separate the fingerprint from the background, remove the background area, determine the main direction and frequency of the fingerprint lines by performing gradient calculation on the fingerprint surface texture grayscale image, separate the adhering fingerprint lines, smooth out the stray point information in the fingerprint image by using an image filtering algorithm, and improve the ridge information of the fingerprint image, and use morphological expansion and corrosion operations to further highlight the undulating characteristics of the fingerprint;

[0128] A feature extraction module, which is used to extract the morphological features of the fingerprint and the local features of the ridges from the preprocessed fingerprint image;

[0129] The similarity comparison module is used to compare the extracted fingerprint features with the fingerprint features stored in the database, use the similarity measurement algorithm to calculate the similarity between the input fingerprint and the fingerprint in the database, and output the user information corresponding to the fingerprint in the database that is higher than the preset similarity value.

[0130] Furthermore, the present solution also proposes a computer-readable storage medium on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned vehicle cabin monitoring method based on dynamic recognition of continuous images is executed.

[0131] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0132] In summary, the advantages of the present invention are: based on the principle of total reflection, it can efficiently and accurately collect the texture image of the fingerprint surface, improve the clarity of fingerprint collection, enhance the reliability of fingerprint recognition, effectively separate the adhered fingerprint lines, avoid recognition errors caused by line adhesion, highlight the undulating characteristics of the fingerprint, enhance the key information in the fingerprint image, improve the accuracy of fingerprint recognition, and can quickly and accurately find the database record that matches the input fingerprint, thereby improving the security of fingerprint recognition.

[0133] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A fingerprint recognition method based on optical recognition sensing technology, characterized in that: include: The acquisition device illuminates the finger through the built-in light source, and based on the different reflectivity of the ridges and valleys of the fingerprint, uses the principle of total reflection of light to collect the fingerprint surface texture image; The collected fingerprint surface texture image is converted into a fingerprint surface texture grayscale image, and the fingerprint is separated from the background to remove the background area; By calculating the gradient of the grayscale image of the fingerprint surface texture, the main direction and frequency of the fingerprint lines are determined, and the adhering fingerprint lines are separated; Use image filtering algorithm to smooth out the stray point information in the fingerprint image and improve the ridge information of the fingerprint image; Morphological dilation and erosion operations are used to further highlight the undulating features of fingerprints; Extracting fingerprint morphological features and local features of ridges from the preprocessed fingerprint image, wherein the fingerprint morphological features include undulation, bifurcation and intersection of the fingerprint, and the local features of ridges include direction, length and spacing of the ridges; Compare the extracted fingerprint features with the fingerprint features stored in the database, use a similarity measurement algorithm to calculate the similarity between the input fingerprint and the fingerprint in the database, and output the user information corresponding to the fingerprint in the database that is higher than the preset similarity value; The method of performing gradient calculation on the grayscale image of the fingerprint surface texture to determine the main direction and frequency of the fingerprint lines and separating the adhering fingerprint lines specifically includes: The horizontal gradient operator is used to perform convolution operation on the grayscale image of the fingerprint surface texture, and the result of the convolution operation is output as the gradient of the grayscale image of the fingerprint surface texture in the x direction; The vertical gradient operator is used to perform convolution operation on the grayscale image of the fingerprint surface texture. The result of the convolution operation is output as the gradient of the grayscale image of the fingerprint surface texture in the y direction. By calculating the square root of the sum of the squares of the gradients in the x-direction and the y-direction, the gradient of each pixel in the grayscale image of the fingerprint surface texture is obtained; By calculating the inverse tangent value of the y-direction gradient and the x-direction gradient, the gradient direction of each pixel in the grayscale image of the fingerprint surface texture is obtained; By sliding a 3×3 window on the fingerprint image, the periodic change of the gradient of each pixel in the window is calculated and output as the frequency; Perform statistical analysis on the directions of all pixels, and output the direction with the highest frequency as the main direction of the fingerprint lines; Based on the main direction of the fingerprint lines, the fingerprint lines in the main direction are enhanced and the adhesion phenomenon in the non-main direction is reduced; The convolution operation is: , In the formula, is the gradient of the pixel at the i-th row and j-th column in the grayscale image of the fingerprint surface texture. When n is x, it is the gradient in the x direction. When n is y, it is the gradient in the y direction. is the gradient operator. When n is x, the gradient operator is , when n is y, the gradient operator is .

2. According to claim 1, a fingerprint recognition method based on optical recognition sensing technology is characterized in that: The step of converting the collected fingerprint surface texture image into a fingerprint surface texture grayscale image and separating the fingerprint from the background, and removing the background area specifically includes: Obtain the weighted coefficients of human eye sensitivity to red, green and blue; Substitute the red, green and blue color components of each pixel in the fingerprint surface texture image into the weighted average formula to obtain the grayscale value of each pixel in the fingerprint surface texture grayscale image; Based on the gray value distribution of the gray image of the fingerprint surface texture, a gray value threshold is preset; Determine whether the grayscale value of each pixel in the grayscale image of the fingerprint surface texture is higher than the grayscale value threshold. If so, mark the pixel as the fingerprint area; if not, mark the pixel as the background area; Remove the background area in the grayscale image of the fingerprint surface texture; The weighted average formula is: , In the formula, is the gray value of the pixel in the i-th row and j-th column in the gray image of the fingerprint surface texture, are the weighted coefficients of human eye sensitivity to red, green and blue, respectively. They are respectively the red, green and blue color components of the pixel at the i-th row and j-th column in the fingerprint surface texture image.

3. The fingerprint recognition method based on optical recognition sensing technology according to claim 2, characterized in that: The method of using an image filtering algorithm to smooth out the stray point information in the fingerprint image and improve the ridge information of the fingerprint image specifically includes: By calculating the average grayscale value of pixels around each pixel in the grayscale image of the fingerprint surface texture, the noise is reduced and the stray point information is filtered out; Based on the main direction of the fingerprint lines, the directional parameter value of the filter response is preset; Based on the spatial frequency of the filter, the frequency parameter value of the filter is preset; Based on the sensitivity to noise and the accuracy requirements of the features, the filter window size parameter value is preset; Preset the phase shift parameter value to adjust the phase response of the filter; Get the filter kernel function formula; The constructed filter is applied to the grayscale image of the fingerprint surface texture. The filter kernel function slides on the image and calculates the convolution result at each position. Output the filtered image; The filter kernel function formula is: ， In the formula, is the filter kernel function, is the directional parameter value of the filter response, is the frequency parameter value of the filter, is the filter window size parameter value, is the phase shift parameter value of the filter.

4. The fingerprint recognition method based on optical recognition sensing technology according to claim 3, characterized in that: The use of morphological dilation and erosion operations to further highlight the undulating features of fingerprints specifically includes: Select structural elements whose size and shape match the ridges and valleys characteristic of the fingerprint; The defined structural element is used to dilate the grayscale image of the fingerprint surface texture, so that the part of the image with brightness higher than the preset brightness threshold expands outward to highlight the convex features of the fingerprint; Check the expanded image to ensure that the ridge line is expanded and avoid image distortion caused by over-expansion; Applying an erosion operation on the dilated image causes the portion of the image with brightness higher than a preset brightness threshold to shrink inward, thus highlighting the concave features of the fingerprint; Adjust the size and shape of the structural element based on the eroded image effect; Check the image after etching to ensure that the valley line is deepened and avoid excessive etching that may cause loss of fingerprint features. Dilation and erosion operations were applied alternately to highlight the undulating features of the fingerprint. After each operation, the image effect was checked and the structural element parameters were adjusted.

5. The fingerprint recognition method based on optical recognition sensing technology according to claim 4, characterized in that: The step of comparing the extracted fingerprint features with the fingerprint features stored in the database, using a similarity measurement algorithm to calculate the similarity between the input fingerprint and the fingerprint in the database, and outputting the user information corresponding to the fingerprint in the database that is higher than a preset similarity value specifically includes: Loading stored fingerprint features and their corresponding user information from the fingerprint database; Convert the extracted fingerprint features and the stored fingerprint features into non-zero feature vector representations; Based on the cosine similarity algorithm, the ratio of the product of the Euclidean norm of the non-zero feature vector converted from the extracted fingerprint feature and the stored fingerprint feature to the dot product between the vectors is calculated and output as the fingerprint similarity; Determine whether the similarity of the fingerprint is higher than the preset similarity threshold. If so, output the user information corresponding to the fingerprint in the database; if not, do not output it.

6. A fingerprint recognition system based on optical recognition sensing technology, used to implement a fingerprint recognition method based on optical recognition sensing technology as claimed in any one of claims 1 to 5, characterized in that: include: An image acquisition module, wherein the image acquisition module is used to collect fingerprint surface texture images by illuminating the fingerprint through a built-in light source and utilizing the principle of total reflection of light based on the different reflectivity of the fingerprint ridges and valleys to light; An image processing module is used to convert the collected fingerprint surface texture image into a fingerprint surface texture grayscale image, separate the fingerprint from the background, remove the background area, determine the main direction and frequency of the fingerprint lines by performing gradient calculation on the fingerprint surface texture grayscale image, separate the adhering fingerprint lines, smooth out the stray point information in the fingerprint image by using an image filtering algorithm, and improve the ridge information of the fingerprint image, and use morphological expansion and corrosion operations to further highlight the undulating characteristics of the fingerprint; A feature extraction module, which is used to extract the morphological features of the fingerprint and the local features of the ridges from the preprocessed fingerprint image; The similarity comparison module is used to compare the extracted fingerprint features with the fingerprint features stored in the database, use the similarity measurement algorithm to calculate the similarity between the input fingerprint and the fingerprint in the database, and output the user information corresponding to the fingerprint in the database that is higher than the preset similarity value.

Images