WO2021004180A1 - 一种纹理特征提取方法、纹理特征提取装置及终端设备 - Google Patents

一种纹理特征提取方法、纹理特征提取装置及终端设备 Download PDF

Info

Publication number
WO2021004180A1
WO2021004180A1 PCT/CN2020/092722 CN2020092722W WO2021004180A1 WO 2021004180 A1 WO2021004180 A1 WO 2021004180A1 CN 2020092722 W CN2020092722 W CN 2020092722W WO 2021004180 A1 WO2021004180 A1 WO 2021004180A1
Authority
WO
WIPO (PCT)
Prior art keywords
pixel
image
curvature
different directions
target
Prior art date
Application number
PCT/CN2020/092722
Other languages
English (en)
French (fr)
Inventor
惠慧
严明洋
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2021004180A1 publication Critical patent/WO2021004180A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/50Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/40Analysis of texture
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12Fingerprints or palmprints
    • G06V40/1347Preprocessing; Feature extraction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20212Image combination
    • G06T2207/20221Image fusion; Image merging

Definitions

  • This application belongs to the field of image processing technology, and in particular relates to a texture feature extraction method, texture feature extraction device and terminal equipment.
  • the palm print contains rich, stable and unique texture feature information, and the palm print collection method is relatively simple. Therefore, palmprint recognition has great application prospects in technologies such as user identity recognition.
  • embodiments of the present application provide a texture feature extraction method, a texture feature extraction device, and a terminal device, which can obtain palmprint texture features with higher definition.
  • the first aspect of the embodiments of the present application provides a texture feature extraction method, including:
  • the grayscale image including palmprint information
  • the grayscale image is filtered in N different directions by a preset filtering method, to obtain filtered images of the grayscale image in the N different directions, and perform filtering processing on the N different directions. Perform fusion processing on the filtered image of to obtain the first image, where N is an integer greater than 0;
  • the target image is obtained by fusing the first image and the second image, and the target image includes the texture feature information in the palmprint information.
  • a second aspect of the embodiments of the present application provides a texture feature extraction device, including:
  • the first acquisition module is configured to acquire a grayscale image, the grayscale image including palmprint information
  • the filtering module is configured to perform filtering processing on the grayscale image in N different directions through a preset filtering method to obtain filtered images of the grayscale image in the N different directions, and Perform fusion processing on N filtered images in different directions to obtain a first image, where N is an integer greater than 0;
  • the second acquisition module is configured to acquire curvature information of each pixel in the grayscale image in M different directions;
  • the first processing module is configured to obtain a second image according to the curvature information of each pixel in the M different directions, where M is an integer greater than 0;
  • the second processing module is configured to fuse the first image and the second image to obtain the target image, and the target image includes texture feature information in the palmprint information.
  • the third aspect of the embodiments of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and running on the processor.
  • the processor executes the computer program
  • the processor may implement the following steps when executing the computer program:
  • the grayscale image including palmprint information
  • the grayscale image is filtered in N different directions by a preset filtering method, to obtain filtered images of the grayscale image in the N different directions, and perform filtering processing on the N different directions. Perform fusion processing on the filtered image of to obtain the first image, where N is an integer greater than 0;
  • the target image is obtained by fusing the first image and the second image, and the target image includes the texture feature information in the palmprint information.
  • the fourth aspect of the embodiments of the present application provides a computer-readable storage medium that stores a computer program that, when executed by a processor, implements the steps of the method described above, for example, When the computer program is executed by the processor, the following steps can be realized:
  • the grayscale image including palmprint information
  • the grayscale image is filtered in N different directions by a preset filtering method, to obtain filtered images of the grayscale image in the N different directions, and perform filtering processing on the N different directions. Perform fusion processing on the filtered image of to obtain the first image, where N is an integer greater than 0;
  • the target image is obtained by fusing the first image and the second image, and the target image includes the texture feature information in the palmprint information.
  • the embodiments of the present application can obtain relatively complete and high-definition palmprint texture characteristics.
  • the palmprint texture information of multiple dimensions is combined, the influence of other interference factors can be reduced, the anti-interference ability is strong, and it has strong practicability and ease of use.
  • FIG. 1 is a schematic diagram of the implementation process of the texture feature extraction method provided by Embodiment 1 of the present application;
  • FIG. 2a is an exemplary schematic diagram of obtaining the pixel value of each pixel in the L-th column of the grayscale image in the vertical direction provided by Embodiment 1 of the present application;
  • 2b is an exemplary schematic diagram of the curve of the pixel value of the pixel point in the Lth column of the grayscale image in the vertical direction and the position sequence of the column provided by Embodiment 1 of the present application
  • Embodiment 3 is a schematic diagram of the implementation process of the texture feature extraction method provided by Embodiment 2 of the present application;
  • FIG. 4 is a schematic diagram of a texture feature extraction device provided by Embodiment 3 of the present application.
  • FIG. 5 is a schematic diagram of a terminal device provided in Embodiment 4 of the present application.
  • the technical solution of the present application can be applied to the field of artificial intelligence, and can analyze palmprints in combination with palmprint texture information of multiple dimensions, so as to obtain relatively complete and high-resolution palmprint texture features.
  • FIG. 1 is a schematic diagram of the implementation process of the texture feature extraction method provided by Embodiment 1 of the present application. As shown in FIG. 1, the texture feature extraction method may include the following steps:
  • Step S101 Obtain a grayscale image, which includes palmprint information.
  • the grayscale image may be a grayscale image obtained after grayscale processing is performed on a non-grayscale image to be processed (such as a red, green, blue (RGB) image, etc.).
  • a non-grayscale image to be processed such as a red, green, blue (RGB) image, etc.
  • RGB red, green, blue
  • other preprocessing such as sharpening, denoising, white balance, etc. can also be performed on the non-grayscale image to be processed.
  • the non-grayscale image to be processed may be an image captured by a device such as a camera. It can be seen that, compared with the devices required for fingerprint identification, the collection equipment and other devices involved in the realization of identity authentication through palmprint recognition are more conventional, and the cost is lower, which is conducive to popularization and application.
  • Step S102 Perform filtering processing on the grayscale image in N different directions by using a preset filtering method to obtain filtered images of the grayscale image in the N different directions respectively, and compare the N The filtered images in different directions are fused to obtain the first image, where N is an integer greater than 0.
  • the filtered images in N different directions may respectively indicate the texture feature information obtained by filtering along the N different directions at the pixel.
  • the preset filtering manner may include a filtering manner capable of extracting texture features.
  • the preset filtering manner may include at least one of filtering manners such as entropy filtering and Gabor filtering.
  • the N may be 1, 2, 4, etc.
  • the N different directions may include one or more of a vertical direction, a horizontal direction, a 45-degree angle direction, and a 135-degree angle direction; for example,
  • the pixel points of the grayscale image can be arranged in a rectangular shape.
  • the grayscale image can also be arranged in a 45-degree angle direction and a 135-degree angle direction to form a pixel column.
  • the length can be different; of course, other directions other than the above directions can also be selected according to the arrangement of the pixels in the array formed by the pixels in the grayscale image.
  • the specific direction selection can be determined according to the selection of N and the actual application scenario, which is not limited here.
  • the preset filtering method may be Gabor filtering.
  • the Gabor filtering may be used to obtain the grayscale image in the vertical direction, the horizontal direction, the 45-degree angle direction, and the The first image is obtained by performing fusion processing on the filtered images in the four directions of the 135-degree angular direction.
  • the direction may indicate the angle of the parallel lines of the Gabor filter; in this case, the direction can be adjusted by adjusting the parameter ⁇ in the Gabor filter.
  • the Gabor filter also includes parameters such as ⁇ wavelength, ⁇ phase shift, ⁇ spatial aspect ratio, and ⁇ bandwidth.
  • the ⁇ indicates the filter scale of the filter; the ⁇ indicates the phase shift of the tuning function; the ⁇ indicates the spatial direction ratio in the Gabor filter, indicating the ellipticity of the Gabor filter; the ⁇ indicates The variance of the Gaussian parameter in the Gabor filter.
  • the Gabor filter also includes the values of ⁇ wavelength, ⁇ phase shift, ⁇ spatial aspect ratio, ⁇ bandwidth and other parameters, which can be set and adjusted according to factors such as application scenarios of the solution.
  • the maximum value of the pixel value of the corresponding pixel in each filtered image may be used as the pixel value of the pixel in the first image, or the pixel value of the corresponding pixel in each filtered image As the pixel value of the pixel in the first image, or the sum of the pixel value of the corresponding pixel in each filtered image is used as the pixel value of the pixel in the first image, so as to realize the The fusion process is described to obtain the first image.
  • the performing fusion processing on the filtered images in the N different directions to obtain the first image includes:
  • the maximum value or the average value of the pixel value of the corresponding pixel in each filtered image may indicate the more prominent texture feature of the pixel in a certain direction obtained through the preset filtering method information.
  • the pixel values of the pixels in the first image can be made to correspond to those in each filtered image.
  • the maximum or average values of the pixel values of the pixels are equal.
  • the pixel value of a pixel located at a certain position may be equal to the maximum or average value of the pixel value of the pixel located at the same position in each filtered image.
  • the correspondence between the pixel points may also have other forms, for example, it may be pixel points at multiple positions in each filtered image, corresponding to a pixel point at one position of the first image.
  • the specific acquisition method of the first image can be set according to actual application scenarios.
  • Step S103 Obtain curvature information of each pixel in the grayscale image in M different directions.
  • the M and the N may be the same or different, which is not limited here; the M different directions may be all the same or partially the same as the N different directions, or may be completely different.
  • the specific direction selection can be determined according to the selection of M and the actual application scenario, which is not limited here.
  • the curvature in the curvature information may be a numerical value of the degree of curvature of the curve at a specified position.
  • the pixels may respectively correspond to different curves, and the different curves may respectively indicate the relationship between the pixel value of the pixel and the position of the pixel in the corresponding direction.
  • the pixel value of each pixel of each column of the grayscale image in that direction can be obtained, and the pixel value of the pixel of each column can be constructed separately A curve with the position sequence of the column, thereby obtaining curvature information of each pixel in the grayscale image in M different directions through the curve.
  • the calculation method of the curvature can be adjusted according to the actual scene.
  • the formula used to calculate the curvature can be:
  • the K is the curvature
  • the y′ is the first derivative corresponding to the pixel.
  • the formula used to calculate the curvature may also be:
  • the K is the curvature
  • the y′′ is the second derivative corresponding to the pixel.
  • Step S104 Obtain a second image according to the curvature information of each pixel in the M different directions, where M is an integer greater than 0.
  • the maximum value of the curvature of each pixel in M different directions may be used as the pixel value of the pixel in the second image, or the average value of the curvature of each pixel in M different directions As the pixel value of the pixel in the second image, or the sum of the curvatures of each pixel in M different directions as the pixel value of the pixel in the second image, etc.
  • Step S105 fusing the first image and the second image to obtain the target image, and the target image includes the texture feature information in the palmprint information.
  • the first image and the second image may be corresponding to the weight, the pixel value of the palmprint texture feature, and/or the pixel value of the background part, etc.
  • An image is merged with the second image to obtain the target image.
  • the palmprint feature of the first image may be low pixel value (such as black), and correspondingly, the background part is high pixel value (such as white); or, it may also be the first image
  • the palmprint feature of is high pixel value (such as white), and correspondingly, the background part is low pixel value (such as black).
  • the palmprint feature may be high pixel value (such as white), and correspondingly, the background part is low pixel value (such as black). Therefore, the specific fusion manner of the first image and the second image may be determined according to the display manner of the first image and the second image. For example, the fusion manner may include combining the first image and the second image. The second image is superimposed or subtracted, etc.
  • the weight can be set according to actual application scenarios. For example, if in a specific application, the texture feature information included in the first image has a high accuracy rate, then the first image can be made to correspond to The weight is higher.
  • normalization processing and/or binarization processing may be performed on the target image to unify the expression mode of the target image.
  • the acquiring the curvature information of each pixel in the grayscale image in M different directions includes:
  • the position sequence may indicate the position information of the pixel in the Lth column of the grayscale image in the corresponding direction.
  • the L-th column of the grayscale image in the corresponding direction includes 5 pixels
  • the position sequence of the column may be a sequence including 1 to 5.
  • the K may be used as an explanatory label without limiting the direction, and the direction K may be specifically determined according to the specific setting forms of the M different directions.
  • the constructing a curve about the pixel value of the pixel point of the column and the position sequence of the column may include: obtaining the corresponding relationship between the pixel value of the pixel point and the position sequence; The corresponding relationship forms a scatter point in a designated coordinate system, and the curve is obtained by fitting according to the scatter point by means of a difference method, calling a designated function, and the like.
  • FIG. 2a an exemplary schematic diagram of obtaining the pixel value of each pixel in the L-th column of the grayscale image in the vertical direction is shown in FIG.
  • obtaining the curvature information of each pixel in the L-th column of the grayscale image in the direction K respectively in the direction includes:
  • each pixel point combination For any one of each pixel point combination, obtain the maximum value of the first curvature of the pixel point contained in the pixel point combination, and set the target curvature of the pixel point contained in the pixel point combination in the direction K Is the product of the maximum value of the first curvature of the pixel points included in the pixel point combination and the number of pixels included in the pixel point combination, and the curvature information includes each pixel point in each pixel point combination The target curvature in the direction K;
  • the obtaining the second image according to the curvature information of each pixel in the M different directions respectively includes:
  • the second image is obtained according to the target curvature of each pixel in the pixel combination in the M different directions.
  • the target second derivative may represent the unevenness of the corresponding pixel in the curve.
  • the target second-order derivative may be the second-order derivative at the corresponding pixel point, or the mean value of the second-order derivative of a plurality of adjacent pixels, and so on.
  • the calculation method of calculating the first curvature of each pixel according to the target second derivative of each pixel can be adjusted according to the actual scene.
  • the formula used to calculate the first curvature may be:
  • the K is the curvature
  • the y' is the first derivative corresponding to the pixel.
  • the formula used to calculate the first curvature may also be:
  • the K is the curvature
  • the y" is the second derivative corresponding to the pixel.
  • the texture feature since the texture feature often has a certain continuity and regionality, the pixel point combination is obtained, and the target curvature of the pixel point contained in the pixel point combination in the direction is Set as the product of the maximum value of the first curvature of the pixel points included in the pixel point combination and the number of pixels included in the pixel point combination, which can be determined and highlighted by the pixel point combination and its value
  • the area where the texture feature is located makes the extracted texture feature information more complete, and makes the main part of the texture feature more prominent, thereby improving the clarity of the obtained palmprint texture feature.
  • obtaining the second image according to the curvature information of each pixel in the M different directions may also include:
  • the pixel value of the pixel point that does not belong to the pixel point combination is set as a preset pixel value.
  • the preset pixel value may be 0; at this time, since the pixel values of the pixel points belonging to the pixel point combination in the second image are generally greater than 0, the pixel points that do not belong to the pixel point The combined pixel points are distinguished from the pixel point combination.
  • the preset pixel value can also take other values.
  • the calculating the target second derivative of each pixel in the Lth column in the direction K according to the curve includes:
  • the preset number is determined by the user according to actual application scenarios, which is ideal, and the preset number may be 1, 2, 3, or 4, etc.
  • the target second derivative can be made more accurate and interference can be reduced.
  • the calculating the first curvature of each pixel according to the target second derivative of each pixel includes:
  • the first curvature of each pixel is calculated based on the first formula, and the first formula is:
  • the K is the first curvature
  • the y" is the second derivative of the target.
  • the first curvature of each pixel is calculated based on the test result of the researcher and the first formula, which can make the texture feature information in the second image more accurate.
  • the gray-scale image is filtered in N different directions through a preset filtering method, and the texture information of the palmprints in the gray-scale image along different directions can be obtained, thereby obtaining more preserved texture information.
  • the first image with the amount of information; at the same time, since the pixel value of the palmprint texture will be significantly different from the non-texture part, the curvature of each pixel in the grayscale image in M different directions is obtained Information
  • the second image is obtained according to the curvature information of each pixel in the M directions, and the second image indicating the change of the pixel value of the pixel in different directions can be obtained; combined with the first
  • the image and the second image can combine palmprint texture information of multiple dimensions to obtain a relatively complete and high-resolution palmprint texture feature.
  • the palmprint texture information of multiple dimensions is combined, the influence of other interference factors can be reduced, the anti-interference ability is strong, and it has strong practicability and ease of use.
  • FIG. 3 is a schematic diagram of the implementation process of the texture feature extraction method provided in the second embodiment of the present application.
  • the texture feature extraction method may include the following steps:
  • Step S301 Obtain a grayscale image, the grayscale image including palmprint information.
  • Step S302 Perform filtering processing on the grayscale image in N different directions by using a preset filtering method, to obtain filtered images of the grayscale image in the N different directions, and compare the N The filtered images in different directions are fused to obtain the first image, where N is an integer greater than 0.
  • Step S303 Obtain curvature information of each pixel in the grayscale image in M different directions.
  • Step S304 Obtain a second image according to the curvature information of each pixel in the M different directions, where M is an integer greater than 0.
  • Steps S301, S302, S303, and S304 in this embodiment are the same as the above-mentioned steps S101, S102, S103, and S104.
  • steps S101, S102, S103, and S104 please refer to related descriptions of steps S101, S102, S103, and S104, which will not be repeated here.
  • Step S305 Obtain respective weights corresponding to the first image and the second image.
  • the weight may be preset.
  • the weight may indicate the definition, completeness, and/or anti-interference degree of the texture features extracted from the first image and the second image in the application.
  • the user can set the weights according to actual application scenarios to meet the needs of different scenarios.
  • Step S306 Perform fusion processing on the first image and the second image according to the weight to obtain a target image, the target image including the texture feature information in the palmprint information.
  • the specific method for obtaining the target image may be determined according to the display method of the first image and the second image.
  • the palmprint feature of the first image may be low pixel value (such as black), and correspondingly, the background part is high pixel value (such as white); or It may be that the palmprint feature of the first image is high pixel value (such as white), and correspondingly, the background part is low pixel value (such as black).
  • the palmprint feature may be high pixel value (such as white), and correspondingly, the background part is low pixel value (such as black).
  • the palmprint feature in the first image may be low pixel value (such as black), and correspondingly, the background part is high pixel value (such as white), while in the second image, so
  • the palmprint feature may be high pixel value (such as white), and correspondingly, the background part is low pixel value (such as black), then the first image and the second image are fused to obtain the target image Specifically, it may include: taking the result of subtracting the pixel values of the corresponding pixels in the second image and the first image as the pixel values of the corresponding pixels in the target image.
  • N filtered images in different directions can be obtained through a preset filtering method (such as Gabor filtering) and the first image is obtained, and the texture in the palmprint is determined to start to appear according to the curvature information of the pixel. And the continuously covered part to obtain a second image; and then perform fusion processing on the first image and the second image according to the weight, thereby combining palmprint texture features of multiple dimensions to obtain a clearer palmprint
  • the target image has texture characteristics, and the weight can be adjusted to meet the needs of multiple different application scenarios in this embodiment.
  • the embodiment of the application does not only obtain the texture feature information of a pixel by adjusting the parameters of a single filter.
  • this embodiment can be used when the amount of calculation is controllable. Next, a clearer texture feature is obtained. In addition, this embodiment has strong anti-interference ability and can reduce the interference of non-palmprint lines.
  • FIG. 4 is a schematic diagram of a texture feature extraction device provided by Embodiment 3 of the present application. For ease of description, only parts related to the embodiments of the present application are shown.
  • the texture feature extraction device 400 includes:
  • the first obtaining module 401 is configured to obtain a grayscale image, where the grayscale image includes palmprint information;
  • the filtering module 402 is configured to perform filtering processing on the grayscale image in N different directions through a preset filtering method, to obtain filtered images of the grayscale image in the N different directions, and Perform fusion processing on the N filtered images in different directions to obtain a first image, where N is an integer greater than 0;
  • the second acquiring module 403 is configured to acquire the curvature information of each pixel in the grayscale image in M different directions;
  • the first processing module 404 is configured to obtain a second image according to the curvature information of each pixel in the M different directions, where M is an integer greater than 0;
  • the second processing module 405 is configured to fuse the first image and the second image to obtain the target image, and the target image includes the texture feature information in the palmprint information.
  • the filtering module 402 is specifically configured to:
  • the second acquiring module 403 specifically includes:
  • the first acquiring unit is used to acquire the pixel value of each pixel in the L-th column of the grayscale image in the direction K, and construct the relationship between the pixel value of the pixel in the L-th column and the column A curve of the position sequence, the curve indicates the change of the pixel value in the column with the position sequence, and L is an integer greater than 0;
  • the first processing unit is configured to obtain curvature information of each pixel in the Lth column of the grayscale image in the direction K according to the curve;
  • the second processing unit is configured to traverse all the columns of the grayscale image in the direction K until the curvature information of all pixels of the grayscale image in the direction K is obtained.
  • the first processing unit specifically includes:
  • the first calculation subunit is configured to calculate the target second derivative of each pixel in the Lth column in the direction K according to the curve;
  • the second calculation subunit is used to calculate the first curvature of each pixel according to the target second derivative of each pixel;
  • a first obtaining subunit configured to obtain a pixel point combination formed by pixels with the first curvature continuously greater than 0 among the pixels in the L th column;
  • the second acquisition subunit is used to acquire the maximum value of the first curvature of the pixel points included in the pixel point combination for any one of the pixel point combinations, and the pixels included in the pixel point combination are in all the pixel points.
  • the target curvature in the direction K is set as the product of the maximum value of the first curvature of the pixel points included in the pixel point combination and the number of pixels included in the pixel point combination, and the curvature information includes each The target curvature of each pixel in the pixel combination in the direction K;
  • the first processing module 404 is specifically configured to:
  • the second image is obtained according to the target curvature of each pixel in the pixel combination in the M different directions.
  • the first calculation subunit is specifically configured to:
  • the second calculation subunit is specifically configured to:
  • the first curvature of each pixel is calculated based on the first formula, and the first formula is:
  • the K is the first curvature
  • the y" is the second derivative of the target.
  • the second processing module 405 specifically includes:
  • the weight unit is used to obtain the weights corresponding to the first image and the second image respectively;
  • the fusion unit is configured to perform fusion processing on the first image and the second image according to the weight to obtain a target image.
  • the gray-scale image is filtered in N different directions through a preset filtering method, and the texture information of the palmprints in the gray-scale image along different directions can be obtained, thereby obtaining more preserved texture information.
  • the first image with the amount of information; at the same time, since the pixel value of the palmprint texture will be significantly different from the non-texture part, the curvature of each pixel in the grayscale image in M different directions is obtained Information
  • the second image is obtained according to the curvature information of each pixel in the M directions, and the second image indicating the change of the pixel value of the pixel in different directions can be obtained; combined with the first
  • the image and the second image can combine palmprint texture information of multiple dimensions to obtain a relatively complete and high-resolution palmprint texture feature.
  • the palmprint texture information of multiple dimensions is combined, the influence of other interference factors can be reduced, the anti-interference ability is strong, and it has strong practicability and ease of use.
  • FIG. 5 is a schematic diagram of a terminal device provided in Embodiment 4 of the present application.
  • the terminal device 5 of this embodiment includes a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and running on the processor 50.
  • the processor 50 executes the computer program 52, the steps in the foregoing embodiments of the texture feature extraction method are implemented, for example, steps 101 to 105 shown in FIG. 1.
  • the processor 50 executes the computer program 52
  • the functions of the modules/units in the foregoing device embodiments such as the functions of the modules 401 to 405 shown in FIG. 4, are realized.
  • the computer program 52 may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 51 and executed by the processor 50 to complete This application.
  • the one or more modules/units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 52 in the terminal device 5.
  • the computer program 52 can be divided into a first acquisition module, a filter module, a second acquisition module, a first processing module, and a second processing module.
  • the specific functions of each module are as follows:
  • the first acquisition module is configured to acquire a grayscale image, the grayscale image including palmprint information
  • the filtering module is configured to perform filtering processing on the grayscale image in N different directions through a preset filtering method to obtain filtered images of the grayscale image in the N different directions, and Perform fusion processing on N filtered images in different directions to obtain a first image, where N is an integer greater than 0;
  • the second acquisition module is configured to acquire curvature information of each pixel in the grayscale image in M different directions;
  • the first processing module is configured to obtain a second image according to the curvature information of each pixel in the M different directions, where M is an integer greater than 0;
  • the second processing module is configured to fuse the first image and the second image to obtain the target image, and the target image includes texture feature information in the palmprint information.
  • the terminal device 5 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
  • the terminal device may include, but is not limited to, a processor 50 and a memory 51.
  • FIG. 5 is only an example of the terminal device 5, and does not constitute a limitation on the terminal device 5. It may include more or less components than shown in the figure, or a combination of certain components, or different components.
  • the terminal device may also include input and output devices, network access devices, buses, etc.
  • the so-called processor 50 may be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the memory 51 may be an internal storage unit of the terminal device 5, such as a hard disk or a memory of the terminal device 5.
  • the memory 51 may also be an external storage device of the terminal device 5, for example, a plug-in hard disk equipped on the terminal device 5, a smart memory card (Smart Media Card, SMC), and a Secure Digital (SD) Card, Flash Card, etc. Further, the memory 51 may also include both an internal storage unit of the terminal device 5 and an external storage device.
  • the memory 51 is used to store the computer program and other programs and data required by the terminal device.
  • the memory 51 can also be used to temporarily store data that has been output or will be output.
  • a computer-readable storage medium stores a computer program, and the computer program implements the steps of the above method when executed by a processor.
  • the computer-readable storage medium may be a non-volatile storage medium or a volatile storage medium.
  • the disclosed apparatus/terminal device and method may be implemented in other ways.
  • the device/terminal device embodiments described above are only illustrative.
  • the division of the modules or units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units.
  • components can be combined or integrated into another system, or some features can be omitted or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • this application implements all or part of the processes in the above-mentioned embodiments and methods, and can also be completed by instructing relevant hardware through a computer program.
  • the computer program can be stored in a computer-readable storage medium. When the program is executed by the processor, the steps of the foregoing method embodiments can be implemented.
  • the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate forms.
  • the computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc.
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • electrical carrier signal telecommunications signal
  • software distribution media etc.
  • the content contained in the computer-readable medium can be appropriately added or deleted in accordance with the requirements of the legislation and patent practice in the jurisdiction.
  • the computer-readable medium Does not include electrical carrier signals and telecommunication signals.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

本申请提供的一种纹理特征提取方法包括:获取灰度图像,所述灰度图像中包括掌纹信息;通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。通过本申请,可以获得清晰度较高的掌纹纹理特征。

Description

一种纹理特征提取方法、纹理特征提取装置及终端设备
本申请要求于2019年7月9日提交中国专利局、申请号为201910614039.5,发明名称为“一种纹理特征提取方法、纹理特征提取装置及终端设备”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请属于图像处理技术领域,尤其涉及一种纹理特征提取方法、纹理特征提取装置及终端设备。
背景技术
掌纹中包含丰富的、具有稳定性和唯一性的纹理特征信息,并且掌纹的采集方式较为简单。因此,在用户身份识别等技术中,掌纹识别具有较大的应用前景。
而发明人意识到,通过传统的使用自适应阈值提取掌纹特征的方法对待识别的掌纹图像的清晰度要求较高,且信息提取的精度较差,因此,在提取掌纹纹理时,往往因为待识别的掌纹图像的清晰度以及该方法本身的限制,而无法最终得到清晰度较高的、较为完整的掌纹纹理。
发明内容
有鉴于此,本申请实施例提供了一种纹理特征提取方法、纹理特征提取装置及终端设备,可以获得清晰度较高的掌纹纹理特征。
本申请实施例的第一方面提供了一种纹理特征提取方法,包括:
获取灰度图像,所述灰度图像中包括掌纹信息;
通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
本申请实施例的第二方面提供了一种纹理特征提取装置,包括:
第一获取模块,用于获取灰度图像,所述灰度图像中包括掌纹信息;
滤波模块,用于通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
第二获取模块,用于获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
第一处理模块,用于根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
第二处理模块,用于将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
本申请实施例的第三方面提供了一种终端设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现如上所述方法的步骤,例如,所述处理器执行所述计算机程序时可实现以下步骤:
获取灰度图像,所述灰度图像中包括掌纹信息;
通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融 合处理,得到第一图像,其中,N为大于0的整数;
获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
本申请实施例的第四方面提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序被处理器执行时实现如上所述方法的步骤,例如,所述计算机程序被处理器执行时可实现以下步骤:
获取灰度图像,所述灰度图像中包括掌纹信息;
通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
本申请实施例能够获得较为完整的、清晰度较高的掌纹纹理特征。此外,本方案中,由于结合了多个维度的掌纹纹理信息,因此可以减少其他干扰因素的影响,抗干扰能力强,具有较强的实用性和易用性。
附图说明
图1是本申请实施例一提供的纹理特征提取方法的实现流程示意图;
图2a是本申请实施例一提供的获取所述灰度图像在竖直方向上的第L列的每一像素点的像素值的一种示例性示意图;
图2b是本申请实施例一提供的关于所述灰度图像在竖直方向上的第L列的像素点的像素值与该列的位置序列的曲线的一种示例性示意图
图3是本申请实施例二提供的纹理特征提取方法的实现流程示意图;
图4是本申请实施例三提供的纹理特征提取装置的示意图;
图5是本申请实施例四提供的终端设备的示意图。
具体实施方式
应当理解,在此本申请说明书中所使用的术语仅仅是出于描述特定实施例的目的而并不意在限制本申请。如在本申请说明书和所附权利要求书中所使用的那样,除非上下文清楚地指明其它情况,否则单数形式的“一”、“一个”及“该”意在包括复数形式。
还应当进一步理解,在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合,并且包括这些组合。
本申请的技术方案可应用于人工智能领域,能够结合多个维度的掌纹纹理信息对掌纹进行分析,以获得较为完整的、清晰度较高的掌纹纹理特征。
图1是本申请实施例一提供的纹理特征提取方法的实现流程示意图,如图1所示该纹理特征提取方法可以包括以下步骤:
步骤S101,获取灰度图像,所述灰度图像中包括掌纹信息。
本申请实施例中,所述灰度图像可以是对待处理的非灰度图像(如红绿蓝(Red Green Blue,RGB)图像等)进行灰度化处理后得到的灰度图像。当然,此外,还可以对所述待处理的非灰度图像做其它预处理,比如锐化、去噪、白平衡等等。
示例性的,所述待处理的非灰度图像可以是通过摄像头等设备拍摄得到的图像。可见,相较于指纹识别所需的装置,通过掌纹识别来实现身份认证时所涉及的采集设备等装置较为常规,成本较低,利于推广应用。
步骤S102,通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数。
本申请实施例中,N个不同方向上的滤波图像可以分别指示该像素点处沿N个不同方向滤波得到的纹理特征信息。所述预设滤波方式可以包括能够进行纹理特征提取的滤波方式。示例性的,所述预设滤波方式可以包括熵滤波(entropy filtering)、Gabor滤波等滤波方式中的至少一种。示例性的,所述N可以取1、2、4等等,所述N个不同方向可以包括竖直方向、水平方向、45度角方向和135度角方向中的一个或多个;例如,所述灰度图像的像素点排列方式可以为矩形,因此,所述灰度图像在45度角方向和135度角方向上,也能够通过像素点排列形成像素列,而此时各个像素列的长度可以不一样;当然,也可以根据所述灰度图像中的像素点所组成的阵列中像素点的排列方式来选择上述方向之外的其它方向。具体的方向选择可以根据所述N的选取以及实际应用场景来确定,在此不作限制。
示例性的,在一个示例中,所述预设滤波方式可以为Gabor滤波,此时,可以通过所述Gabor滤波,获得所述灰度图像分别在竖直方向、水平方向、45度角方向和135度角方向这四个方向上的滤波图像,并将四个所述滤波图像进行融合处理,得到所述第一图像。所述方向可以指示所述Gabor滤波器的平行线的角度;此时,通过调整所述Gabor滤波器中的参数θ即可调整所述方向。
此外,需要说明的是,所述Gabor滤波器中还包括λ波长、ψ相位偏移、γ空间纵横比、σ带宽等参数。所述λ指示滤波器的滤波尺度;所述ψ指示调谐函数的相位偏移;所述γ指示所述Gabor滤波器中的空间方向比例,指示所述Gabor滤波器的椭圆率;所述σ指示所述Gabor滤波器中高斯参数的方差。所述Gabor滤波器中还包括λ波长、ψ相位偏移、γ空间纵横比、σ带宽等参数的取值,可以根据本方案的应用场景等因素进行设置和调整。
本申请实施例中,可以将各个滤波图像中对应的像素点的像素值的最大值作为所述第一图像中所述像素点的像素值,或者将各个滤波图像中对应的像素点的像素值的均值作为所述第一图像中所述像素点的像素值,或者将各个滤波图像中对应的像素点的像素值的和作为所述第一图像中所述像素点的像素值,从而实现所述融合处理,以得到第一图像。
可选的,所述对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,包括:
获取各个滤波图像中相对应的像素点的像素值的最大值或者均值,并根据所述相对应的像素点的像素值的最大值或者均值,获得第一图像。
本申请实施例中,所述各个滤波图像中对应的像素点的像素值的最大值或者均值可以指示通过所述预设滤波方式所获取到的、该像素点在某一方向上较为显著的纹理特征信息。
本申请实施例中,通过根据所述相对应的像素点的像素值的最大值或者均值获得第一图像,可以使得所述第一图像中的像素点的像素值与各个滤波图像中相对应的像素点的像素值的最大值或者均值相等。例如,第一图像中,位于某一位置处的像素点的像素值,可以与各个滤波图像中位于相同位置处的像素点的像素值的最大值或者均值相等。当然,所述像素点之间的对应关系也可以有其他形式,例如,可以是各个滤波图像中的多个位置处的像素点,对应于第一图像的一个位置处的像素点。所述第一图像的具体获取方式可以根据实际应用场景需要进行设置。
步骤S103,获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息。
本申请实施例中,所述M和所述N可以相同,也可以不同,在此不作限制;所述M 个不同方向可以与上述的N个不同方向全部相同或部分相同,也可以完全不同,具体的方向选择可以根据所述M的选取以及实际应用场景来确定,在此不作限制。
所述曲率信息中的曲率可以为曲线在指定位置处的弯曲程度的数值。相应地,在所述M个不同方向上,所述像素点可以分别对应不同的曲线,所述不同的曲线可以分别指示在对应方向上所述像素点的像素值随像素点的位置的变化关系。例如,对于所述M个不同方向中的任意一个方向,可以获取所述灰度图像在该方向上的每一列的每一像素点的像素值,并分别构建关于每一列的像素点的像素值与该列的位置序列的曲线,从而通过所述曲线获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息。需要说明的是,本申请实施例中,所述曲率的计算方式可以根据实际场景进行调整。一般地,计算所述曲率所依据的公式可以为:
Figure PCTCN2020092722-appb-000001
其中,所述K为曲率,所述y′为所述像素点所对应的一阶导数。而本申请实施例中,根据实际测试的结果,计算所述曲率所依据的公式也可以为:
Figure PCTCN2020092722-appb-000002
其中,所述K为曲率,所述y″为所述像素点所对应的二阶导数。
步骤S104,根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数。
本申请实施例中,所述根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像的具体实现方式可以有多种。例如,可以将各个像素点分别在M个不同方向上的曲率的最大值作为所述第二图像中所述像素点的像素值,或者将各个像素点分别在M个不同方向上的曲率的均值作为所述第二图像中所述像素点的像素值,或者将各个像素点分别在M个不同方向上的曲率的和作为所述第二图像中所述像素点的像素值等等。
步骤S105,将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
本申请实施例中,示例性的,可以根据所述第一图像和所述第二图像分别对应的权重、掌纹纹理特征的像素值和/或背景部分的像素值等信息,将所述第一图像与所述第二图像进行融合来获得所述目标图像。例如,由于Gabor滤波器中参数选取不同,第一图像的掌纹特征可能是低像素值(如黑色),相应的,背景部分是高像素值(如白色);或者,也有可能是第一图像的掌纹特征为高像素值(如白色),相应的,背景部分是低像素值(如黑色)。而第二图像中,所述掌纹特征可能是高像素值(如白色),相应的,背景部分是低像素值(如黑色)。因此,可以根据所述第一图像和第二图像的显示方式确定所述第一图像与所述第二图像的具体融合方式,例如,所述融合方式可以包括将所述第一图像与所述第二图像相叠加或者相减等等。并且,所述权重可以根据实际的应用场景进行设置,例如若在某一具体应用场合中,所述第一图像中包括的纹理特征信息准确率较高,则可以使得所述第一图像所对应的权重较高。
可选的,本申请实施例中,在获得所述目标图像之后,还可以对所述目标图像做归一化处理和/或二值化处理,以统一所述目标图像的表示方式。
可选的,所述获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息, 包括:
对于所述M个不同方向中的任意一个方向K:
获取所述灰度图像在所述方向K上的第L列的每一像素点的像素值,并构建关于所述第L列的像素点的像素值与该列的位置序列的曲线,所述曲线指示该列中所述像素值随位置序列的变化情况,L为大于0的整数;
根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息;
遍历所述灰度图像在所述方向K上的所有列,直到获得所述灰度图像的所有像素点在所述方向K上的曲率信息。
本申请实施例中,所述位置序列可以指示所述灰度图像在对应方向上的第L列的像素点位置信息。例如,所述灰度图像在对应方向上的第L列包括5个像素点,则该列的位置序列可以为包括1到5的序列。本实施例中,所述K可以作为说明性质的标号,而不对所述方向进行限制,所述方向K具体可以根据所述M个不同方向的具体设置形式来确定。
在一个实施方式中,具体地,所述构建关于该列的像素点的像素值与该列的位置序列的曲线可以包括:获取所述像素点的像素值与所述位置序列的对应关系;根据所述对应关系在指定坐标系中形成散点,并根据所述散点,通过差值法、调用指定函数等方式拟合得到所述曲线。
示例性的,如图2a所示获取所述灰度图像在竖直方向上的第L列的每一像素点的像素值的示例性示意图,如图2b所示为关于该列的像素点的像素值与该列的位置序列的曲线的一种示例性示意图。
可选的,所述根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息,包括:
根据所述曲线,计算所述方向K上的第L列的每一像素点的目标二阶导数;
分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率;
获取所述第L列的像素点中,所述第一曲率连续大于0的像素点所构成的像素点组合;
对于各个像素点组合中的任意一个,获取该像素点组合中所包含的像素点的第一曲率的最大值,将该像素点组合中所包含的像素点在所述方向K上的目标曲率设置为该像素点组合中所包含的像素点的第一曲率的最大值与该像素点组合所包含的像素点的个数的乘积,并使所述曲率信息包含各个像素点组合中的各个像素点在所述方向K上的目标曲率;
所述根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,包括:
根据所述像素点组合中的各个像素点分别在所述M个不同方向上的目标曲率,得到所述第二图像。
本申请实施例中,所述目标二阶导数可以表现所述曲线中,对应的像素点处的凹凸性。所述目标二阶导数可以是对应的像素点处的二阶导数,也可以是相邻的多个像素点的二阶导数的均值等等。
所述分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率的计算方式可以根据实际场景进行调整。一般地,计算所述第一曲率所依据的公式可以为:
Figure PCTCN2020092722-appb-000003
其中,所述K为曲率,所述y’为所述像素点所对应的一阶导数。或者,基于实际测试结果,计算所述第一曲率所依据的公式也可以为:
Figure PCTCN2020092722-appb-000004
其中,所述K为曲率,所述y”为所述像素点所对应的二阶导数。
本申请实施例中,由于所述纹理特征往往具有一定的连续性、区域性,因此通过获取所述像素点组合,并将所述像素点组合中所包含的像素点在该方向上的目标曲率设置为该像素点组合中所包含的像素点的第一曲率的最大值与该像素点组合所包含的像素点的个数的乘积,可以通过所述像素点组合及其取值确定并突出连续的纹理特征所在的区域,使得提取的纹理特征信息更为完整,并使得纹理特征的主体部分较为突出,从而提高了获取到的掌纹纹理特征的清晰度。
此外,可选的,所述根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,还可以包括:
将所述第二图像中,不属于像素点组合的像素点的像素值设置为预设像素值。
示例性的,本申请实施例中,所述预设像素值可以为0;此时,由于第二图像中属于像素点组合的像素点的像素值一般都大于0,因此可以将不属于像素点组合的像素点与像素点组合区分开来。当然,所述预设像素值也可以取其他值。
其中,可选的,所述根据所述曲线,计算所述方向K上第L列的每一像素点的目标二阶导数,包括:
对于所述方向K上第L列的各个像素点中的任意一个,获取该像素点以及该像素点在所述方向K的第L列上预设个数的相邻像素点的二阶导数的均值,并以所述均值作为所述像素点的目标二阶导数。
本申请实施例中,所述预设个数是用户根据实际应用场景进行确定,是理想的,所述预设个数可以是1、2、3或4等。通过取均值,可以使所述目标二阶导数更为准确,减少干扰。
其中,可选的,所述分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率,包括:
根据各个像素点的目标二阶导数,基于第一公式计算各个像素点的第一曲率,所述第一公式为:
Figure PCTCN2020092722-appb-000005
其中,所述K为所述第一曲率,所述y"为所述目标二阶导数。
本申请实施例中,基于研发人员测试结果,基于第一公式计算各个像素点的第一曲率,可以使得到的第二图像中的纹理特征信息更为准确。
本申请实施例通过预设滤波方式在N个不同方向上对所述灰度图像进行滤波处理,可以获取到所述灰度图像中的掌纹沿不同方向的纹理信息,从而获得保留了较多信息量的第一图像;同时,由于所述掌纹纹理的像素值会与非纹理部分存在明显差异,因此,通过获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息,根据各个像素点分别在所述M个方向上的曲率信息,得到所述第二图像,可以获取到指示所述像素点的像素值沿不同方向的变化情况的第二图像;结合第一图像与第二图像,可以结合多个维度的掌纹纹理信息,获得较为完整的、清晰度较高的掌纹纹理特征。此外,本方案中,由于结合了多个维度的掌纹纹理信息,因此可以减少其他干扰因素的影响,抗干扰能力强,具有较强的实用性和易用性。
在上述实施例的基础上,图3是本申请实施例二提供的纹理特征提取方法的实现流程示意图,如图3所示该纹理特征提取方法可以包括以下步骤:
步骤S301,获取灰度图像,所述灰度图像中包括掌纹信息。
步骤S302,通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理, 获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数。
步骤S303,获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息。
步骤S304,根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数。
本实施例步骤S301、S302、S303、S304与上述步骤S101、S102、S103、S104相同,具体可参见步骤S101、S102、S103、S104相关描述,在此不再赘述。
步骤S305,获取所述第一图像和第二图像所分别对应的权重。
本申请实施例中,所述权重可以是预先设置的。所述权重可以指示在应用中,所述第一图像和所述第二图像所分别提取到的纹理特征的清晰度、完整程度和/或抗干扰程度等。用户可以根据实际应用场景来设置所述权重,以适应不同场景的需要。
步骤S306,根据所述权重,对所述第一图像与所述第二图像进行融合处理,获得目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
本申请实施例中,可以根据所述第一图像和第二图像的显示方式确定获得目标图像的具体方式。例如,由于预设滤波方式中的滤波器的参数选取不同,第一图像的掌纹特征可能是低像素值(如黑色),相应的,背景部分是高像素值(如白色);或者,也有可能是第一图像的掌纹特征为高像素值(如白色),相应的,背景部分是低像素值(如黑色)。而第二图像中,所述掌纹特征可能是高像素值(如白色),相应的,背景部分是低像素值(如黑色)。示例性的,在实际应用中,所述第一图像中的掌纹特征可能是低像素值(如黑色),相应的,背景部分是高像素值(如白色),而第二图像中,所述掌纹特征可能是高像素值(如白色),相应的,背景部分是低像素值(如黑色),则所述对所述第一图像与所述第二图像进行融合处理,获得目标图像具体可以包括:将所述第二图像与所述第一图像中对应的像素点的像素值相减的结果作为目标图像中对应像素点的像素值。
本申请实施例中,可以通过预设滤波方式(如Gabor滤波)获得N个不同方向的滤波图像并得到第一图像,并根据所述像素点的曲率信息来确定所述掌纹中纹理开始出现以及连续覆盖的部分,得到第二图像;再根据所述权重对所述第一图像与所述第二图像进行融合处理,从而结合多个维度的掌纹纹理特征,得到较为清晰的包含掌纹纹理特征的目标图像,并且,本实施例可以通过调节所述权重来适应多种不同应用场景的需要。本申请实施例不仅仅通过调节单一的滤波器的参数来获取像素点的纹理特征信息,因此,相较于仅采用诸如Gabor滤波的纹理特征提取方法,本实施例可以在计算量可控的情况下,获得较为清晰的纹理特征。并且,本实施例抗干扰能力强,能够减少非掌纹纹路的干扰。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
图4是本申请实施例三提供的纹理特征提取装置的示意图。为了便于说明,仅示出了与本申请实施例相关的部分。
所述纹理特征提取装置400包括:
第一获取模块401,用于获取灰度图像,所述灰度图像中包括掌纹信息;
滤波模块402,用于通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
第二获取模块403,用于获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
第一处理模块404,用于根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
第二处理模块405,用于将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
可选的,所述滤波模块402具体用于:
获取各个滤波图像中相对应的像素点的像素值的最大值或者均值,并根据所述相对应的像素点的像素值的最大值或者均值,获得第一图像。
可选的,对于所述M个不同方向中的任意一个方向K,所述第二获取模块403具体包括:
第一获取单元,用于获取所述灰度图像在所述方向K上的第L列的每一像素点的像素值,并构建关于所述第L列的像素点的像素值与该列的位置序列的曲线,所述曲线指示该列中所述像素值随位置序列的变化情况,L为大于0的整数;
第一处理单元,用于根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息;
第二处理单元,用于遍历所述灰度图像在所述方向K上的所有列,直到获得所述灰度图像的所有像素点在所述方向K上的曲率信息。
可选的,所述第一处理单元具体包括:
第一计算子单元,用于根据所述曲线,计算所述方向K上的第L列的每一像素点的目标二阶导数;
第二计算子单元,用于分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率;
第一获取子单元,用于获取所述第L列的像素点中,所述第一曲率连续大于0的像素点所构成的像素点组合;
第二获取子单元,用于对于各个像素点组合中的任意一个,获取该像素点组合中所包含的像素点的第一曲率的最大值,将该像素点组合中所包含的像素点在所述方向K上的目标曲率设置为该像素点组合中所包含的像素点的第一曲率的最大值与该像素点组合所包含的像素点的个数的乘积,并使所述曲率信息包含各个像素点组合中的各个像素点在所述方向K上的目标曲率;
所述第一处理模块404具体用于:
根据所述像素点组合中的各个像素点分别在所述M个不同方向上的目标曲率,得到所述第二图像。
可选的,所述第一计算子单元具体用于:
对于所述方向K上第L列的各个像素点中的任意一个,获取该像素点以及该像素点在所述方向K的第L列上预设个数的相邻像素点的二阶导数的均值,并以所述均值作为所述像素点的目标二阶导数。
可选的,所述第二计算子单元具体用于:
根据各个像素点的目标二阶导数,基于第一公式计算各个像素点的第一曲率,所述第一公式为:
Figure PCTCN2020092722-appb-000006
其中,所述K为所述第一曲率,所述y"为所述目标二阶导数。
可选的,所述第二处理模块405具体包括:
权重单元,用于获取所述第一图像和第二图像所分别对应的权重;
融合单元,用于根据所述权重,对所述第一图像与所述第二图像进行融合处理,获得目标图像。
本申请实施例通过预设滤波方式在N个不同方向上对所述灰度图像进行滤波处理,可以获取到所述灰度图像中的掌纹沿不同方向的纹理信息,从而获得保留了较多信息量的第一图像;同时,由于所述掌纹纹理的像素值会与非纹理部分存在明显差异,因此,通过获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息,根据各个像素点分别在所述M个方向上的曲率信息,得到所述第二图像,可以获取到指示所述像素点的像素值沿不同方向的变化情况的第二图像;结合第一图像与第二图像,可以结合多个维度的掌纹纹理信息,获得较为完整的、清晰度较高的掌纹纹理特征。此外,本方案中,由于结合了多个维度的掌纹纹理信息,因此可以减少其他干扰因素的影响,抗干扰能力强,具有较强的实用性和易用性。
图5是本申请实施例四提供的终端设备的示意图。如图5所示,该实施例的终端设备5包括:处理器50、存储器51以及存储在所述存储器51中并可在所述处理器50上运行的计算机程序52。所述处理器50执行所述计算机程序52时实现上述各个纹理特征提取方法实施例中的步骤,例如图1所示的步骤101至105。或者,所述处理器50执行所述计算机程序52时实现上述各装置实施例中各模块/单元的功能,例如图4所示模块401至405的功能。
示例性的,所述计算机程序52可以被分割成一个或多个模块/单元,所述一个或者多个模块/单元被存储在所述存储器51中,并由所述处理器50执行,以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段,该指令段用于描述所述计算机程序52在所述终端设备5中的执行过程。例如,所述计算机程序52可以被分割成第一获取模块、滤波模块、第二获取模块、第一处理模块、第二处理模块,各模块具体功能如下:
第一获取模块,用于获取灰度图像,所述灰度图像中包括掌纹信息;
滤波模块,用于通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
第二获取模块,用于获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
第一处理模块,用于根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
第二处理模块,用于将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
所述终端设备5可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述终端设备可包括,但不仅限于,处理器50、存储器51。本领域技术人员可以理解,图5仅仅是终端设备5的示例,并不构成对终端设备5的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件,例如所述终端设备还可以包括输入输出设备、网络接入设备、总线等。
所称处理器50可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器51可以是所述终端设备5的内部存储单元,例如终端设备5的硬盘或内存。所述存储器51也可以是所述终端设备5的外部存储设备,例如所述终端设备5上配备 的插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)等。进一步地,所述存储器51还可以既包括所述终端设备5的内部存储单元也包括外部存储设备。所述存储器51用于存储所述计算机程序以及所述终端设备所需的其他程序和数据。所述存储器51还可以用于暂时地存储已经输出或者将要输出的数据。
所属领域的技术人员可以清楚地了解到,为了描述的方便和简洁,仅以上述各功能单元、模块的划分进行举例说明,实际应用中,可以根据需要而将上述功能分配由不同的功能单元、模块完成,即将所述装置的内部结构划分成不同的功能单元或模块,以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中,上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。另外,各功能单元、模块的具体名称也只是为了便于相互区分,并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请的示例性实施例中,还提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序被处理器执行时实现如上所述方法的步骤。可选的,该计算机可读存储介质可以是非易失性的存储介质,也可以是易失性的存储介质。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述或记载的部分,可以参见其它实施例的相关描述。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
在本申请所提供的实施例中,应该理解到,所揭露的装置/终端设备和方法,可以通过其它的方式实现。例如,以上所描述的装置/终端设备实施例仅仅是示意性的,例如,所述模块或单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口,装置或单元的间接耦合或通讯连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请实现上述实施例方法中的全部或部分流程,也可以通过计算机程序来指令相关的硬件来完成,所述的计算机程序可存储于一计算机可读存储介质中,该计算机程序在被处理器执行时,可实现上述各个方法实施例的步骤。其中,所述计算机程序包括计算机程序代码,所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-Only Memory)、随机存取存 储器(RAM,Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,计算机可读介质不包括电载波信号和电信信号。
以上所述实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述实施例对本申请进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围,均应包含在本申请的保护范围之内。

Claims (20)

  1. 一种纹理特征提取方法,其中,包括:
    获取灰度图像,所述灰度图像中包括掌纹信息;
    通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
    获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
    根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
    将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
  2. 如权利要求1所述的纹理特征提取方法,其中,所述对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,包括:
    获取各个滤波图像中相对应的像素点的像素值的最大值或者均值,并根据所述相对应的像素点的像素值的最大值或者均值,获得第一图像。
  3. 如权利要求1所述的纹理特征提取方法,其中,所述获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息,包括:
    对于所述M个不同方向中的任意一个方向K:
    获取所述灰度图像在所述方向K上的第L列的每一像素点的像素值,并构建关于所述第L列的像素点的像素值与该列的位置序列的曲线,所述曲线指示该列中所述像素值随位置序列的变化情况,L为大于0的整数;
    根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息;
    遍历所述灰度图像在所述方向K上的所有列,直到获得所述灰度图像的所有像素点在所述方向K上的曲率信息。
  4. 如权利要求3所述的纹理特征提取方法,其中,所述根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息,包括:
    根据所述曲线,计算所述方向K上的第L列的每一像素点的目标二阶导数;
    分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率;
    获取所述第L列的像素点中,所述第一曲率连续大于0的像素点所构成的像素点组合;
    对于各个像素点组合中的任意一个,获取该像素点组合中所包含的像素点的第一曲率的最大值,将该像素点组合中所包含的像素点在所述方向K上的目标曲率设置为该像素点组合中所包含的像素点的第一曲率的最大值与该像素点组合所包含的像素点的个数的乘积,并使所述曲率信息包含各个像素点组合中的各个像素点在所述方向K上的目标曲率;
    所述根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,包括:
    根据所述像素点组合中的各个像素点分别在所述M个不同方向上的目标曲率,得到所述第二图像。
  5. 如权利要求4所述的纹理特征提取方法,其中,所述根据所述曲线,计算所述方向K上第L列的每一像素点的目标二阶导数,包括:
    对于所述方向K上第L列的各个像素点中的任意一个,获取该像素点以及该像素点在所述方向K的第L列上预设个数的相邻像素点的二阶导数的均值,并以所述均值作为所述像素点的目标二阶导数。
  6. 如权利要求4所述的纹理特征提取方法,其中,所述分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率,包括:
    根据各个像素点的目标二阶导数,基于第一公式计算各个像素点的第一曲率,所述第一公式为:
    Figure PCTCN2020092722-appb-100001
    其中,所述K为所述第一曲率,所述y"为所述目标二阶导数。
  7. 如权利要求1至6任意一项所述的纹理特征提取方法,其中,所述将所述第一图像与所述第二图像进行融合获得所述目标图像,包括:
    获取所述第一图像和第二图像所分别对应的权重;
    根据所述权重,对所述第一图像与所述第二图像进行融合处理,获得目标图像。
  8. 一种纹理特征提取装置,其中,包括:
    第一获取模块,用于获取灰度图像,所述灰度图像中包括掌纹信息;
    滤波模块,用于通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
    第二获取模块,用于获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
    第一处理模块,用于根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
    第二处理模块,用于将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
  9. 一种终端设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序,其中,所述处理器执行所述计算机程序时实现以下步骤:
    获取灰度图像,所述灰度图像中包括掌纹信息;
    通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
    获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
    根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
    将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
  10. 如权利要求9所述的终端设备,其中,所述处理器在执行所述对所述N个不同方向上的滤波图像进行融合处理,得到第一图像时,具体执行以下步骤:
    获取各个滤波图像中相对应的像素点的像素值的最大值或者均值,并根据所述相对应的像素点的像素值的最大值或者均值,获得第一图像。
  11. 如权利要求9所述的终端设备,其中,所述处理器在执行所述获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息时,具体执行以下步骤:
    对于所述M个不同方向中的任意一个方向K:
    获取所述灰度图像在所述方向K上的第L列的每一像素点的像素值,并构建关于所述第L列的像素点的像素值与该列的位置序列的曲线,所述曲线指示该列中所述像素值随位置序列的变化情况,L为大于0的整数;
    根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息;
    遍历所述灰度图像在所述方向K上的所有列,直到获得所述灰度图像的所有像素点在所述方向K上的曲率信息。
  12. 如权利要求11所述的终端设备,其中,所述处理器在执行所述根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息时,具体执行以下步骤:
    根据所述曲线,计算所述方向K上的第L列的每一像素点的目标二阶导数;
    分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率;
    获取所述第L列的像素点中,所述第一曲率连续大于0的像素点所构成的像素点组合;
    对于各个像素点组合中的任意一个,获取该像素点组合中所包含的像素点的第一曲率的最大值,将该像素点组合中所包含的像素点在所述方向K上的目标曲率设置为该像素点组合中所包含的像素点的第一曲率的最大值与该像素点组合所包含的像素点的个数的乘积,并使所述曲率信息包含各个像素点组合中的各个像素点在所述方向K上的目标曲率;
    所述处理器在执行所述根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像时,具体执行以下步骤:
    根据所述像素点组合中的各个像素点分别在所述M个不同方向上的目标曲率,得到所述第二图像。
  13. 如权利要求12所述的终端设备,其中,所述处理器在执行所述分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率时,具体执行以下步骤:
    根据各个像素点的目标二阶导数,基于第一公式计算各个像素点的第一曲率,所述第一公式为:
    Figure PCTCN2020092722-appb-100002
    其中,所述K为所述第一曲率,所述y"为所述目标二阶导数。
  14. 如权利要求12所述的终端设备,其中,所述处理器在执行所述根据所述曲线,计算所述方向K上第L列的每一像素点的目标二阶导数时,具体执行以下步骤:
    对于所述方向K上第L列的各个像素点中的任意一个,获取该像素点以及该像素点在所述方向K的第L列上预设个数的相邻像素点的二阶导数的均值,并以所述均值作为所述像素点的目标二阶导数。
  15. 如权利要求9至14任意一项所述的终端设备,其中,所述处理器在执行所述将所述第一图像与所述第二图像进行融合获得所述目标图像时,具体执行以下步骤:
    获取所述第一图像和第二图像所分别对应的权重;
    根据所述权重,对所述第一图像与所述第二图像进行融合处理,获得目标图像。
  16. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其中,所述计算机程序被处理器执行时实现以下步骤:
    获取灰度图像,所述灰度图像中包括掌纹信息;
    通过预设滤波方式,在N个不同方向上对所述灰度图像进行滤波处理,获得所述灰度图像分别在所述N个不同方向上的滤波图像,并对所述N个不同方向上的滤波图像进行融合处理,得到第一图像,其中,N为大于0的整数;
    获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息;
    根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,其中,M为大于0的整数;
    将所述第一图像与所述第二图像进行融合获得所述目标图像,所述目标图像中包括所述掌纹信息中的纹理特征信息。
  17. 如权利要求16所述的计算机可读存储介质,其中,所述对所述N个不同方向上的滤波图像进行融合处理,得到第一图像时,所述计算机程序被处理器执行以具体实现以下步骤:
    获取各个滤波图像中相对应的像素点的像素值的最大值或者均值,并根据所述相对应的像素点的像素值的最大值或者均值,获得第一图像。
  18. 如权利要求16所述的计算机可读存储介质,其中,所述获取所述灰度图像中的各个像素点分别在M个不同方向上的曲率信息时,所述计算机程序被处理器执行以具体实现以下步骤:
    对于所述M个不同方向中的任意一个方向K:
    获取所述灰度图像在所述方向K上的第L列的每一像素点的像素值,并构建关于所述第L列的像素点的像素值与该列的位置序列的曲线,所述曲线指示该列中所述像素值随位置序列的变化情况,L为大于0的整数;
    根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息;
    遍历所述灰度图像在所述方向K上的所有列,直到获得所述灰度图像的所有像素点在所述方向K上的曲率信息。
  19. 如权利要求18所述的计算机可读存储介质,其中,所述根据所述曲线,获得所述灰度图像在所述方向K上的第L列的各个像素点分别在该方向上的曲率信息时,所述计算机程序被处理器执行以具体实现以下步骤:
    根据所述曲线,计算所述方向K上的第L列的每一像素点的目标二阶导数;
    分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率;
    获取所述第L列的像素点中,所述第一曲率连续大于0的像素点所构成的像素点组合;
    对于各个像素点组合中的任意一个,获取该像素点组合中所包含的像素点的第一曲率的最大值,将该像素点组合中所包含的像素点在所述方向K上的目标曲率设置为该像素点组合中所包含的像素点的第一曲率的最大值与该像素点组合所包含的像素点的个数的乘积,并使所述曲率信息包含各个像素点组合中的各个像素点在所述方向K上的目标曲率;
    所述根据各个像素点分别在所述M个不同方向上的曲率信息,得到第二图像,包括:
    根据所述像素点组合中的各个像素点分别在所述M个不同方向上的目标曲率,得到所述第二图像。
  20. 如权利要求19所述的计算机可读存储介质,其中,所述分别根据各个像素点的目标二阶导数计算各个像素点的第一曲率时,所述计算机程序被处理器执行以具体实现以下步骤:
    根据各个像素点的目标二阶导数,基于第一公式计算各个像素点的第一曲率,所述第一公式为:
    Figure PCTCN2020092722-appb-100003
    其中,所述K为所述第一曲率,所述y"为所述目标二阶导数。
PCT/CN2020/092722 2019-07-09 2020-05-27 一种纹理特征提取方法、纹理特征提取装置及终端设备 WO2021004180A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910614039.5 2019-07-09
CN201910614039.5A CN110473242B (zh) 2019-07-09 2019-07-09 一种纹理特征提取方法、纹理特征提取装置及终端设备

Publications (1)

Publication Number Publication Date
WO2021004180A1 true WO2021004180A1 (zh) 2021-01-14

Family

ID=68507587

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/092722 WO2021004180A1 (zh) 2019-07-09 2020-05-27 一种纹理特征提取方法、纹理特征提取装置及终端设备

Country Status (2)

Country Link
CN (1) CN110473242B (zh)
WO (1) WO2021004180A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113077472A (zh) * 2021-04-07 2021-07-06 华南理工大学 一种纸质心电图曲线图像分割方法、系统、装置及介质
CN113240595A (zh) * 2021-05-06 2021-08-10 腾讯科技(深圳)有限公司 图像检测方法、装置、存储介质及计算机设备
CN113822804A (zh) * 2021-09-23 2021-12-21 深圳万兴软件有限公司 灰度图的优化方法、装置、计算机设备及存储介质
CN114581376A (zh) * 2022-01-31 2022-06-03 南通摩瑞纺织有限公司 基于图像识别的纺织用蚕茧自动拣选方法及系统
CN114739315A (zh) * 2022-03-09 2022-07-12 上海颢汉数字技术有限公司 彩塑扫描速率控制方法、装置、电子设备及存储介质
CN115019159A (zh) * 2022-08-09 2022-09-06 济宁安泰矿山设备制造有限公司 一种泵轴承故障快速识别方法
CN115240013A (zh) * 2022-09-20 2022-10-25 惠民县黄河先进技术研究院 基于图像数据分析的织物肌理自动分类方法及系统
CN116030058A (zh) * 2023-03-29 2023-04-28 无锡斯达新能源科技股份有限公司 一种用于抛光垫表面粗糙度的质量评价方法
CN117853484A (zh) * 2024-03-05 2024-04-09 湖南建工交建宏特科技有限公司 一种基于视觉的桥梁损伤智能监测方法及系统
CN114739315B (zh) * 2022-03-09 2024-05-31 上海颢汉数字技术有限公司 彩塑扫描速率控制方法、装置、电子设备及存储介质

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110473242B (zh) * 2019-07-09 2022-05-27 平安科技(深圳)有限公司 一种纹理特征提取方法、纹理特征提取装置及终端设备
CN111398292A (zh) * 2020-04-07 2020-07-10 苏州哈工吉乐优智能装备科技有限公司 一种基于gabor滤波和CNN的布匹瑕疵检测方法、系统及设备
CN111445421A (zh) * 2020-04-08 2020-07-24 Oppo广东移动通信有限公司 图像滤波方法、装置、存储介质及移动设备
CN113470155B (zh) * 2021-09-03 2021-11-23 贝壳技术有限公司 纹理图像处理方法和装置、电子设备和存储介质
CN113674272B (zh) * 2021-09-06 2024-03-15 上海集成电路装备材料产业创新中心有限公司 图像检测方法与装置
CN115690246B (zh) * 2022-10-20 2023-06-27 北京国电通网络技术有限公司 图像纹理信息生成方法、装置、设备、介质和程序产品

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090220127A1 (en) * 2008-02-28 2009-09-03 Honeywell International Inc. Covariance based face association
CN101533475A (zh) * 2009-04-08 2009-09-16 华南师范大学 一种基于形状自适应邻域的遥感图像特征提取方法
US20120314913A1 (en) * 2011-06-08 2012-12-13 Honeywell International Inc. System and method for ocular recognition
CN106022218A (zh) * 2016-05-06 2016-10-12 浙江工业大学 一种基于小波变换和Gabor滤波器的掌纹掌静脉图像层融合方法
CN107346434A (zh) * 2017-05-03 2017-11-14 上海大学 一种基于多特征及支持向量机的植物病虫害检测方法
CN110473242A (zh) * 2019-07-09 2019-11-19 平安科技(深圳)有限公司 一种纹理特征提取方法、纹理特征提取装置及终端设备

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102264163B1 (ko) * 2014-10-21 2021-06-11 삼성전자주식회사 텍스쳐를 처리하는 방법 및 장치
CN109902586A (zh) * 2019-01-29 2019-06-18 平安科技(深圳)有限公司 掌纹提取方法、装置及存储介质、服务器

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090220127A1 (en) * 2008-02-28 2009-09-03 Honeywell International Inc. Covariance based face association
CN101533475A (zh) * 2009-04-08 2009-09-16 华南师范大学 一种基于形状自适应邻域的遥感图像特征提取方法
US20120314913A1 (en) * 2011-06-08 2012-12-13 Honeywell International Inc. System and method for ocular recognition
CN106022218A (zh) * 2016-05-06 2016-10-12 浙江工业大学 一种基于小波变换和Gabor滤波器的掌纹掌静脉图像层融合方法
CN107346434A (zh) * 2017-05-03 2017-11-14 上海大学 一种基于多特征及支持向量机的植物病虫害检测方法
CN110473242A (zh) * 2019-07-09 2019-11-19 平安科技(深圳)有限公司 一种纹理特征提取方法、纹理特征提取装置及终端设备

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
RAO QING: "Research of Finger Vein Recognition Algorithms Based on Location and Direction Coding", CHINESE MASTER'S THESES FULL-TEXT DATABASE, no. 7, 1 May 2013 (2013-05-01), pages 1 - 80, XP055772673 *
WU; JIACUN: "Finger Vein Recognition Algorithm Research", MASTER THESIS, no. 5, 1 May 2019 (2019-05-01), pages 1 - 69, XP009525513 *

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113077472A (zh) * 2021-04-07 2021-07-06 华南理工大学 一种纸质心电图曲线图像分割方法、系统、装置及介质
CN113240595B (zh) * 2021-05-06 2023-09-08 腾讯科技(深圳)有限公司 图像检测方法、装置、存储介质及计算机设备
CN113240595A (zh) * 2021-05-06 2021-08-10 腾讯科技(深圳)有限公司 图像检测方法、装置、存储介质及计算机设备
CN113822804A (zh) * 2021-09-23 2021-12-21 深圳万兴软件有限公司 灰度图的优化方法、装置、计算机设备及存储介质
CN114581376A (zh) * 2022-01-31 2022-06-03 南通摩瑞纺织有限公司 基于图像识别的纺织用蚕茧自动拣选方法及系统
CN114581376B (zh) * 2022-01-31 2023-03-24 南通摩瑞纺织有限公司 基于图像识别的纺织用蚕茧自动拣选方法及系统
CN114739315A (zh) * 2022-03-09 2022-07-12 上海颢汉数字技术有限公司 彩塑扫描速率控制方法、装置、电子设备及存储介质
CN114739315B (zh) * 2022-03-09 2024-05-31 上海颢汉数字技术有限公司 彩塑扫描速率控制方法、装置、电子设备及存储介质
CN115019159A (zh) * 2022-08-09 2022-09-06 济宁安泰矿山设备制造有限公司 一种泵轴承故障快速识别方法
CN115019159B (zh) * 2022-08-09 2022-10-25 济宁安泰矿山设备制造有限公司 一种泵轴承故障快速识别方法
CN115240013A (zh) * 2022-09-20 2022-10-25 惠民县黄河先进技术研究院 基于图像数据分析的织物肌理自动分类方法及系统
CN116030058A (zh) * 2023-03-29 2023-04-28 无锡斯达新能源科技股份有限公司 一种用于抛光垫表面粗糙度的质量评价方法
CN117853484A (zh) * 2024-03-05 2024-04-09 湖南建工交建宏特科技有限公司 一种基于视觉的桥梁损伤智能监测方法及系统
CN117853484B (zh) * 2024-03-05 2024-05-28 湖南建工交建宏特科技有限公司 一种基于视觉的桥梁损伤智能监测方法及系统

Also Published As

Publication number Publication date
CN110473242A (zh) 2019-11-19
CN110473242B (zh) 2022-05-27

Similar Documents

Publication Publication Date Title
WO2021004180A1 (zh) 一种纹理特征提取方法、纹理特征提取装置及终端设备
US11120254B2 (en) Methods and apparatuses for determining hand three-dimensional data
CN111488756B (zh) 基于面部识别的活体检测的方法、电子设备和存储介质
CN111815754B (zh) 一种三维信息确定方法、三维信息确定装置及终端设备
CN109117773B (zh) 一种图像特征点检测方法、终端设备及存储介质
CN109871845B (zh) 证件图像提取方法及终端设备
US11455831B2 (en) Method and apparatus for face classification
CN109948439B (zh) 一种活体检测方法、系统及终端设备
CN111080526A (zh) 航拍图像的农田面积的测算方法、装置、设备及介质
CN112241933A (zh) 人脸图像处理方法、装置、存储介质及电子设备
CN107908998B (zh) 二维码解码方法、装置、终端设备及计算机可读存储介质
CN112991180A (zh) 图像拼接方法、装置、设备以及存储介质
WO2022233185A1 (zh) 一种图像滤波方法、装置、终端和计算机可读存储介质
CN107610097A (zh) 仪表定位方法、装置和终端设备
CN113393540A (zh) 图像中的色边像素点的确定方法、装置和计算机设备
CN109117693B (zh) 一种基于广角取景的扫描识别的方法及终端
CN111383254A (zh) 一种深度信息获取方法、系统及终端设备
CN111199169A (zh) 图像处理方法和装置
CN111126250A (zh) 一种基于ptgan的行人重识别方法及装置
WO2022199395A1 (zh) 人脸活体检测方法、终端设备及计算机可读存储介质
CN111222446A (zh) 人脸识别方法、人脸识别装置及移动终端
CN109286798B (zh) 一种投影屏幕的边框位置识别方法、系统及终端设备
CN111786992A (zh) 会员注册和认证方法、终端设备及可读存储介质
CN111354058A (zh) 一种图像着色方法、装置、图像采集设备及可读存储介质
US20220245819A1 (en) Method for processing images, electronic device, and storage medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20836852

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20836852

Country of ref document: EP

Kind code of ref document: A1