WO2016127735A1 - 纹线距离的计算方法和装置 - Google Patents

纹线距离的计算方法和装置 Download PDF

Info

Publication number
WO2016127735A1
WO2016127735A1 PCT/CN2016/070192 CN2016070192W WO2016127735A1 WO 2016127735 A1 WO2016127735 A1 WO 2016127735A1 CN 2016070192 W CN2016070192 W CN 2016070192W WO 2016127735 A1 WO2016127735 A1 WO 2016127735A1
Authority
WO
WIPO (PCT)
Prior art keywords
pixel value
pixel
boundary point
pixels
adjacent
Prior art date
Application number
PCT/CN2016/070192
Other languages
English (en)
French (fr)
Inventor
郑利
徐坤平
杨云
Original Assignee
比亚迪股份有限公司
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 比亚迪股份有限公司 filed Critical 比亚迪股份有限公司
Priority to JP2017542060A priority Critical patent/JP6494776B2/ja
Priority to EP16748536.6A priority patent/EP3264361A4/en
Priority to KR1020177022541A priority patent/KR101985689B1/ko
Priority to US15/548,688 priority patent/US20180018497A1/en
Publication of WO2016127735A1 publication Critical patent/WO2016127735A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/13—Sensors therefor
    • G06V40/1318—Sensors therefor using electro-optical elements or layers, e.g. electroluminescent sensing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1347—Preprocessing; Feature extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1347—Preprocessing; Feature extraction
    • G06V40/1359—Extracting features related to ridge properties; Determining the fingerprint type, e.g. whorl or loop
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/14—Measuring arrangements characterised by the use of optical techniques for measuring distance or clearance between spaced objects or spaced apertures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20021—Dividing image into blocks, subimages or windows
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person

Definitions

  • the present invention relates to the field of image processing technologies, and in particular, to a method and an apparatus for calculating a line distance.
  • biometric-based identity authentication methods are easy to forget, leak, loss, forgery, etc., causing inconvenience and security problems for life.
  • biometric-based identity authentication methods fingerprint recognition technology is the earliest and most widely used one. It has the characteristics of high stability, uniqueness, easy collection and high security. Therefore, fingerprint is an ideal biometric for identity authentication, and the market share of fingerprint recognition has also increased year by year.
  • the fingerprint recognition system since the fingerprint image belongs to personal privacy, the fingerprint recognition system generally does not directly store the image of the fingerprint, but extracts the feature information of the fingerprint from the fingerprint image through an algorithm, and then performs fingerprint matching and identification to complete the identity authentication. Therefore, the high reliability fingerprint recognition algorithm is the key to ensure the correct identification of fingerprints.
  • the line distance is defined as the distance between a given ridge line and an adjacent valley line.
  • the length of the center of the ridge line to the center of the valley line is calculated as the line distance.
  • the larger the line distance the more sparse the line is. On the contrary, the smaller the line distance, the denser the line.
  • the size of the line distance is determined by the structure of the fingerprint itself and the resolution of the image acquisition.
  • the pixel gradation value exhibits a feature of a discrete sinusoidal waveform, as shown in FIG. 1, the distance between the two ridge lines can represent Is the distance between the peak and the peak in a sinusoidal waveform.
  • the noise information mainly comes from the sensor itself and the actual situation of the finger with water, oil, peeling, etc., resulting in a sinusoidal extreme value situation, such as: can not have a single peak, in fact can not be accurate Find this extreme point.
  • the same fingerprint is pressed with the same force, and the collected fingerprint image has a large difference in the distance of the line obtained by this method at the same position of the fingerprint.
  • the distribution of the ridges and valley lines on the fingerprint along its perpendicular to the direction of the ridge is not an ideal sinusoidal waveform, and there is no spike. Highlighting the peaks, therefore, the grayscale-based line distance algorithm can only accommodate clear and uniform fingerprint images.
  • an object of the present invention is to provide a method for calculating a line distance, which improves the line distance by finding the boundary point of the fingerprint ridge line and the valley line, and calculating the line distance according to the coordinates of the boundary point and the sub-pixel value. Accuracy and anti-noise ability, more accurately reflect the global density of fingerprints, and a wider range of applications.
  • a second object of the present invention is to provide a device for calculating a ridge distance.
  • a method for calculating a ridge distance includes the steps of: acquiring an original image and performing gradation processing to generate a grayscale image; generating a normal map and cutting according to the grayscale image.
  • the map determines a normal direction of a center point of each of the partitions; traversing pixels in a normal direction of a center point of each of the partitions in each of the partitions to calculate each of the partitions
  • the apparatus for calculating the line distance finds the boundary point of the fingerprint ridge line and the valley line, and calculates the line distance according to the coordinates of the boundary point and the sub-pixel value, so that the line distance is more accurate and closer to the fingerprint.
  • FIG. 1 is a schematic view showing a sinusoidal distribution characteristic of a local region of a ridge in the related art
  • FIG. 2 is a flow chart of a method of calculating a ridge distance according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a boundary point position and a number of changes of a pixel value from 0 to 1 according to an embodiment of the present invention
  • FIG. 4 is a flow chart for calculating sub-pixel values of boundary points in accordance with one embodiment of the present invention.
  • 5A-5C are schematic diagrams of calculating sub-pixel values along a horizontal direction, in accordance with one embodiment of the present invention.
  • 6A-6C are schematic diagrams of calculating sub-pixel values along a non-vertical or horizontal direction, in accordance with one embodiment of the present invention.
  • FIG. 7A is a schematic diagram of a grayscale image in accordance with one embodiment of the present invention.
  • Figure 7B is a schematic illustration of a tangential view in accordance with one embodiment of the present invention.
  • FIG. 7C is a schematic diagram of a smoothed image in accordance with one embodiment of the present invention.
  • 7D is a schematic diagram of a binary image according to an embodiment of the present invention.
  • 7E is a diagram showing the result of Gabor filtering on the data of the final per-line line distance according to an embodiment of the present invention.
  • the present invention proposes a method and a device for calculating the ridge distance.
  • a method and apparatus for calculating the ridge distance of the embodiment of the present invention will be described below with reference to the accompanying drawings.
  • FIG. 2 is a flow chart of a method of calculating a ridge distance in accordance with one embodiment of the present invention. As shown in FIG. 2, the method for calculating the line distance according to the embodiment of the present invention includes the following steps:
  • the original image is acquired and subjected to gradation processing to generate a grayscale image.
  • the original image is subjected to gradation processing to generate a grayscale image A(i, j).
  • the normal map O1(i, j) and the tangential map O2(i, j) can be obtained from the grayscale image A(i, j) by the gradient method.
  • filtering the grayscale image according to the tangential map to generate the smoothed image comprises: performing 1*7 mean filtering on the grayscale image according to the tangential map to generate a smoothed image.
  • the gradation map O2(i, j) is used to perform 1*7 mean filtering deburring on the grayscale image A(i, j) to obtain a smoothed image B(i, j), which is then smoothed by differential binarization.
  • the image B(i,j) is converted into a binary image D(i,j).
  • the binary image D(i,j) is divided into blocks of N*N (for example, N is 33), wherein the blocks are swiped point by point, so there is overlap between the blocks and the blocks.
  • the first pixel value is zero and the second pixel value is one.
  • the first pixel value is zero and the second pixel value is one.
  • each block of the binary image D(i,j) in each block, the pixels in the normal direction of each block center point are traversed, and the calculation of the adjacent two pixels is performed.
  • the pixel value changes from 0 to 1 and the number of changes from 1 to 0 and the pixel coordinates at the change (ie, the coordinates of the boundary point).
  • the striped area represents the ridge line
  • the pixel value is 0, and the blank area is In the valley line
  • the pixel value is 1
  • the position indicated by the arrow is the position from 0 to 1, that is, the position of the boundary point, and the number of changes from 0 to 1 in the diagram 3 is 3 times.
  • the sub-pixel values of the corresponding boundary points at the change are generated by the following steps:
  • the sub-pixel boundary is calculated while traversing, and is calculated in two cases, one is to calculate sub-pixels of the boundary point along the oblique direction, and the other is to calculate in the vertical or horizontal direction. The following two cases are explained separately.
  • the preset direction is a vertical direction; when the normal angle corresponding to the normal direction of the center point of the block is equal to 90 degrees, the preset direction is a horizontal direction; and the normal direction at the center point of the block corresponds to
  • the angle is equal to 0 degrees or 90, if the pixel values of two pixels adjacent to the boundary point are the same as the pixel values of the boundary point, the sub-pixel value of the boundary point is 0; if two adjacent to the boundary point If only one of the pixel values of the pixel is the same as the pixel value of the boundary point, the sub-pixel value of the boundary point is 0.5; if the pixel values of the two pixels adjacent to the boundary point are different from the pixel values of the boundary point, the boundary point The subpixel value is 1.
  • FIGS. 5A to 5C a schematic diagram of calculating sub-pixel values along the horizontal direction (in the y-axis direction in FIG. 5A), wherein, in the present invention, a fingerprint image (as shown in FIG. 7A)
  • the upper left corner of the graph is the origin, and the vertical and horizontal boundaries of the fingerprint image establish a coordinate system for the x-axis and the y-axis, respectively, and the normal angle corresponding to the normal direction of the center point of the partition is the center of the block The angle between the normal direction of the point and the x-axis.
  • the vertical stripe fill block represents the boundary point
  • the white fill block represents the point on the valley line
  • the black filled block represents the point on the ridge line.
  • the pixel value of the two pixels on both sides of the boundary point is used for judging, and if both are the same as the pixel value of the boundary point, the ⁇ value (sub-pixel of the boundary point) is 0. If there is a pixel value equal to the boundary point, ⁇ is 1/2; if none of the pixel values of the boundary point are the same, ⁇ takes a value of 1.
  • two pixels adjacent to the boundary point in FIG. 5A are white filled blocks
  • two pixels adjacent to the boundary point in FIG. 5B are black and white filled blocks, respectively, and two adjacent to the boundary points in FIG. 5C
  • Each pixel is a black filled block.
  • FIGS. 6A to 6C a schematic diagram of calculating sub-pixel values in a non-vertical or horizontal direction (ie, an oblique direction, as shown in FIG. 6A), in which a vertical stripe-filled block represents a boundary Point, the white filled block represents the point on the valley line, and the black filled block represents the point on the ridge line.
  • a vertical stripe-filled block represents a boundary Point
  • the white filled block represents the point on the valley line
  • the black filled block represents the point on the ridge line.
  • the ⁇ value is 0; the pixel having one pixel The value is the same as the pixel value of the boundary point, and the ⁇ value is 1/4; if the pixel values of the adjacent two pixels are not the same as the pixel value of the boundary point, the ⁇ value is 1/2.
  • two pixels adjacent to the boundary point are white filled blocks
  • two pixels adjacent to the boundary point in FIG. 6B are black and white filled blocks, respectively, and two adjacent to the boundary point in FIG. 6C.
  • Each pixel is a black filled block.
  • the ridge distance of the center point of a block is generated by the following formula:
  • num1 and num2 are the number of changes of the pixel value between the first pixel value and the pixel value between the second pixel values
  • num1 is the pixel value of the adjacent two pixels in the block changes from the second pixel value to the first
  • num2 is the number of times the pixel value of two adjacent pixels in the block changes from the first pixel value to the second pixel value
  • X 1 and X num1 are respectively the inner edge of the block
  • the pixel direction of the adjacent two pixels in the normal direction of the center point of the block changes from the second pixel value to the first pixel value and the pixel value of the adjacent two pixels from the second pixel value from the second pixel value from the second pixel value
  • the change is the abscissa value of the corresponding boundary point at the first pixel value
  • Y 1 and Y num2 are the pixel values of the adjacent two pixels appearing in the normal direction along the center point of the block in the block, respectively.
  • ⁇ is the point method to block the center point angle ranges from 0 to ⁇
  • ⁇ Xi is the direction in the sub-block
  • the center point of the method occurs in a direction the pixel values of the i-th two adjacent second pixel value from the change value to sub-pixel corresponding to a first boundary point at a pixel value
  • ⁇ Yi along the inside of the block In the normal direction of the center point of the block, the pixel value of the adjacent two pixels appears from the first pixel value to the sub-pixel value of the corresponding boundary point at the second pixel value, D1(i, j) and D2 (i, j) respectively calculating the pixel value according to the adjacent two pixels from the second pixel value to the first pixel value and calculating the pixel value according to the adjacent two pixels from the first pixel value
  • the method further includes: acquiring, according to the number of times the pixel values of two adjacent pixels in each of the blocks change between the first pixel value and the second pixel value, a boundary within each block The number of points; if the number of boundary points in the block is smaller than the preset number of blocks, the pixels in the opposite direction of the normal direction of the center point of the block are further traversed in the block.
  • D1(i,j) and D2(i,j) need to be satisfied: 1D1, D2 In each block, there must be two or more boundary points from pixel value 0 to pixel value 1 or pixel value 1 to pixel value 0 value conversion, that is, a line distance operation requires two The ridge line and a valley line are completed, or by two valley lines and one ridge line. If not, D1 or D2 does not exist; 2 if one of D1 or D2 does not exist, the other D1 or D2 needs to find at least 2 converted boundary points in the opposite direction of the normal direction.
  • the method for calculating the ridge distance of the embodiment of the present invention finds the boundary point of the fingerprint ridge line and the valley line, and calculates the ridge line distance according to the coordinates of the boundary point and the sub-pixel value, so that the ridge line distance is more accurate and closer to the fingerprint reality.
  • the feature thus more accurately reflects the global density feature of the fingerprint, and the method has strong anti-noise ability and wider application range.
  • the method further comprises: performing 5*5 local area mean filtering on the line distance.
  • 5*5 local area mean filtering is performed on the calculated line distance to play a smoothing effect, and the final line distance per point is obtained.
  • FIG. 7A is a schematic diagram of the gray image
  • FIG. 7B Shown as a schematic diagram of a tangential diagram
  • FIG. 7C is a schematic diagram of a smooth image
  • FIG. 7D is a schematic diagram of a binary image
  • FIG. 7E is a schematic diagram of a result of Gabor filtering on data of a final line distance per point. .
  • the method for calculating the ridge distance of the embodiment of the present invention avoids the situation that the extreme value of the fingerprint sinusoidal curve is relatively complicated in the related art, for example, where there is a theoretical maximum point, there are more than two uncertainties.
  • the maximum value of the number cannot accurately determine the fingerprint line distance.
  • the method in the embodiment of the present invention finds the boundary point of the fingerprint ridge line and the valley line, and determines that there is only one point, and an indefinite number of boundary points does not appear.
  • the image with noise is highly redundant, and the condition requirements are not critical, which expands the scope of application.
  • the method has high engineering application value and can provide reliable parameters for later image filtering, segmentation, ridge tracking and matching.
  • an embodiment of the present invention further provides a slanting distance calculation device, which is provided by the stencil distance calculation device provided by the embodiment of the present invention.
  • the calculation method of the ridge distance provided by the several embodiments corresponds to the calculation method of the ridge distance, and the method for calculating the ridge distance provided by the embodiment is not detailed in this embodiment. description.
  • FIG. 8 is a schematic structural diagram of a device for calculating a line distance according to an embodiment of the present invention. As shown in FIG.
  • the apparatus for calculating the ridge distance of the embodiment of the present invention includes: a gradation processing module 100, a generation module 200, a smoothing processing module 300, a blocking processing module 400, a sub-pixel calculation module 500, and a line distance.
  • a module 600 is generated.
  • the grayscale processing module 100 is configured to acquire an original image and perform grayscale processing to generate a grayscale image.
  • the generating module 200 is configured to generate a normal map and a tangential map according to the grayscale image.
  • the smoothing processing module 300 is configured to filter the grayscale image according to the tangential map to generate a smoothed image, and convert the smoothed image into a binary image.
  • the smoothing processing module 300 is specifically configured to: perform 1*7 mean filtering on the grayscale image according to the tangential map to generate a smoothed image, and convert the smoothed image into a binary image.
  • the blocking processing module 400 is configured to block the binary image and determine the normal direction of the center point of each of the partitions according to the normal map.
  • the sub-pixel calculation module 500 is configured to traverse the pixels in the normal direction of the center point of each of the blocks in each of the blocks to calculate the pixel values of the adjacent two pixels in each of the blocks.
  • the first pixel value is zero and the second pixel value is one.
  • the sub-pixel calculation module 500 is further configured to: obtain, according to the number of times the pixel values of two adjacent pixels in each block change between the first pixel value and the second pixel value.
  • the number of boundary points in each block, and the number of boundary points in the block is smaller than the preset number of blocks, and the opposite of the normal direction of the center point of the block in the block The upward pixels are traversed.
  • the sub-pixel calculation module 500 generates a sub-pixel value of the boundary point, specifically: acquiring pixel values of two pixels adjacent to the boundary point along the preset direction, and according to the boundary point The pixel values of the two adjacent pixels and the pixel values of the boundary points calculate the sub-pixel values of the boundary points.
  • the sub-pixel calculation module 500 determines that the preset direction is a vertical direction when the normal angle corresponding to the normal direction of the center point of the block is equal to 0 degrees; at the center point of the block When the normal angle corresponding to the normal direction is equal to 90 degrees, the preset direction is determined to be a horizontal direction; and is further used to generate when the pixel values of two pixels adjacent to the boundary point are the same as the pixel values of the boundary point.
  • the sub-pixel value of the boundary point is 0; when only one of the pixel values of the two pixels adjacent to the boundary point is the same as the pixel value of the boundary point, the sub-pixel value of the generated boundary point is 0.5; adjacent to the boundary point When the pixel values of the two pixels are different from the pixel values of the boundary points, the sub-pixel value of the generated boundary point is 1.
  • the sub-pixel calculation module 500 determines that the preset direction is apart from the vertical direction when the normal angle corresponding to the normal direction of the center point of the block is not equal to 0 degrees and not equal to 90 degrees. And a direction other than the horizontal direction; and further, when the pixel values of the two pixels adjacent to the boundary point are the same as the pixel values of the boundary point, the sub-pixel value of the generated boundary point is 0; When only one of the pixel values of the adjacent two pixels is the same as the pixel value of the boundary point, the sub-pixel value of the generated boundary point is 0.25; the pixel values of the two pixels adjacent to the boundary point are the pixel values of the boundary point At the same time, the sub-pixel value of the generated boundary point is 0.5.
  • the ridge distance generation module 600 is configured to change the coordinates and sub-pixel values of the corresponding boundary points according to the number of times and the change between the first pixel value and the second pixel value of the pixel values of adjacent two pixels in each of the blocks. Generate a line distance.
  • the ridge distance generation module 600 generates ridge distances by equations (1), (2), and (3).
  • the ridge distance generation module 600 is further configured to perform 5*5 local area averaging filtering on the ridge distance.
  • the device for calculating the line distance of the embodiment of the present invention calculates the line boundary of the fingerprint ridge line and the valley line, and calculates the line distance according to the coordinates of the boundary point and the sub-pixel value, so that the line distance is more accurate and closer to the fingerprint reality.
  • first and second are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated.
  • features defining “first” or “second” may include at least one of the features, either explicitly or implicitly.
  • the meaning of "a plurality” is at least two, such as two, three, etc., unless specifically defined otherwise.
  • a "computer-readable medium” can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with the instruction execution system, apparatus, or device.
  • computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example, by paper or other medium, followed by editing, solution The program is processed electronically in other suitable ways, if necessary, and then stored in computer memory.
  • portions of the invention may be implemented in hardware, software, firmware or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: having logic gates for implementing logic functions on data signals. Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.
  • each functional unit in each embodiment of the present invention may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module.
  • the above integrated modules can be implemented in the form of hardware or in the form of software functional modules.
  • the integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.
  • the above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

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

Abstract

一种纹线距离的计算方法和装置，该方法包括：获取原始图像并进行灰度处理以生成灰度图像（S1）；根据灰度图像生成法向图和切向图（S2）；根据切向图对灰度图像进行滤波以生成平滑图像，并将平滑图像转换为二值图像（S3）；对二值图像进行分块，并根据法向图确定每个分块的中心点的法向方向（S4）；在每个分块内对在每个分块的中心点的法向方向上的像素进行遍历以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数、变化处对应边界点的坐标和亚像素值，第一像素值为脊线所在的像素的像素值，第二像素值为谷线所在的像素的像素值（S5）；根据变化的次数和变化处对应边界点的坐标和亚像素值生成纹线距离（S6）。该方法通过寻找脊线和谷线的边界点并根据边界点的坐标和亚像素值计算纹线距离，提高了准确度，且抗噪声能力强，适用范围更广。

Description

纹线距离的计算方法和装置 技术领域
本发明涉及图像处理技术领域，尤其涉及一种纹线距离的计算方法和装置。
背景技术
随着社会的发展，人们对身份认证的准确性、安全性与实用性提出了更高的要求。传统的身份认证方式如密码，钥匙、身份证等，存在容易忘记，泄露，丢失，伪造等问题，给生活带来不便和安全问题。基于生物识别技术的身份认证，可以克服传统身份认证的许多缺点，目前已经成为安全技术研究的热点。在各种基于生物特征的身份认证方法中，指纹识别技术是应用最早，最广泛的一种。其具有高稳定性，独特性，易采集，安全性高等特点，因此，指纹是一种较理想的可用于身份认证的生物特征，指纹识别的市场份额也逐年攀升。由于指纹图像属于个人隐私，指纹识别系统一般并不直接存储指纹的图像，而是通过算法从指纹图像中提取指纹的特征信息，再进行指纹匹配识别，完成身份认证。因此，高可靠性的指纹识别算法是保证正确辨识指纹的关键。
其中，纹线距离的定义为：给定脊线和相邻谷线的距离。一般情况下，通过计算脊线中心到谷线中心的长度作为纹线距离。纹线距离越大，表明该处纹线越稀疏；反之，纹线距离越小，表明该处纹线越密集。纹线距离的大小决定于指纹本身的结构和图像采集的分辨率。
相关技术中关于指纹纹线距离的算法大致可以分为两类：第一类是基于整幅图像的指纹纹线距离估计，理想地认为一幅指纹图像纹线距离分布是正态的，但在实际指纹库中，同一枚指纹纹线距离会出现二倍的差异，故纹线距离不可以基于整幅图像计算；第二类是基于图像区域的局部纹线距离估计，需要准确地找到频谱的峰值点，这在算法上是很难做到的，求出的纹线距离会不准确。
例如，在相关技术中的第二类算法中，在指纹图像垂直纹线方向上，像素灰度值呈现离散正弦波形的特征，如附图1所示，两条脊线之间的距离可以表示为正弦波形中波峰与波峰之间的距离。由于传感器实际采集的指纹图像，会含有噪声，噪声信息主要来自传感器本身以及手指有水、油、脱皮等实际情况，导致正弦曲线极值情况比较复杂，如：不能有单一峰值，实际上不能准确找到这个极值点。同一枚指纹用相同力度按压，采集到的指纹图片，指纹同一位置用此方法求得的纹线距离有较大差异。对于指纹灰度图像本身来说，指纹上的脊线、谷线沿着其垂直于纹线方向的分布并不是理想的正弦波形，也没有尖峰的 突出峰值，因此，基于灰度的纹线距离算法只能适应清晰均匀的指纹图像。
发明内容
本发明旨在至少在一定程度上解决相关技术中的技术问题之一。为此，本发明的一个目的在于提出一种纹线距离的计算方法，该方法通过寻找指纹脊线和谷线的边界点，并根据边界点的坐标和亚像素值计算纹线距离，提高了准确度，且抗噪声能力强，更准确的反映了指纹全局疏密特征，适用范围更广。
本发明的第二个目的在于提出一种纹线距离的计算装置。
为了实现上述目的，本发明第一方面实施例的纹线距离的计算方法，包括以下步骤：获取原始图像并进行灰度处理以生成灰度图像；根据所述灰度图像生成法向图和切向图；根据所述切向图对所述灰度图像进行滤波以生成平滑图像，并将所述平滑图像转换为二值图像；对所述二值图像进行分块，并根据所述法向图确定每个分块的中心点的法向方向；在所述每个分块内对在所述每个分块的中心点的法向方向上的像素进行遍历，以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数、变化处对应边界点的坐标和亚像素值，其中，所述第一像素值为脊线所在的像素的像素值，所述第二像素值为谷线所在的像素的像素值；以及根据所述每个分块内的所述相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数和所述变化处对应边界点的坐标和亚像素值生成纹线距离。
根据本发明实施例的纹线距离的计算方法，通过寻找指纹脊线和谷线的边界点，并根据边界点的坐标和亚像素值计算纹线距离，使得纹线距离更加准确，更接近指纹真实特征，从而更准确的反映了指纹全局疏密特征，且该方法的抗噪声能力强，适用范围更广。
为了实现上述目的，本发明第二方面实施例的纹线距离的计算装置，包括：灰度处理模块，用于获取原始图像并进行灰度处理以生成灰度图像；生成模块，用于根据所述灰度图像生成法向图和切向图；平滑处理模块，用于根据所述切向图对所述灰度图像进行滤波以生成平滑图像，并将所述平滑图像转换为二值图像；分块处理模块，用于对所述二值图像进行分块，并根据所述法向图确定每个分块的中心点的法向方向；亚像素计算模块，用于在所述每个分块内对在所述每个分块的中心点的法向方向上的像素进行遍历，以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数、变化处对应边界点的坐标和亚像素值，其中，所述第一像素值为脊线所在的像素的像素值，所述第二像素值为谷线所在的像素的像素值；以及纹线距离生成模块，用于根据所述每个分块内的所述相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数和所述变化处对应边界点的坐标和亚像素值生成纹线距离。
根据本发明实施例的纹线距离的计算装置，通过寻找指纹脊线和谷线的边界点，并根据边界点的坐标和亚像素值计算纹线距离，使得纹线距离更加准确，更接近指纹真实特征，从而更准确的反映了指纹全局疏密特征，而且抗噪声能力强，适用范围更广。
附图说明
图1是相关技术中纹线局部区域正弦分布特性的示意图；
图2是根据本发明一个实施例的纹线距离的计算方法的流程图；
图3是根据本发明一个实施例的像素值由0变为1的边界点位置以及变化次数的示意图；
图4是根据本发明一个实施例的计算边界点的亚像素值的流程图；
图5A-图5C是根据本发明一个实施例的沿着水平方向计算亚像素值的示意图；
图6A-图6C是根据本发明一个实施例的沿着非竖直或者水平方向计算亚像素值的示意图；
图7A是根据本发明一个实施例的灰度图像的示意图；
图7B是根据本发明一个实施例的切向图的示意图；
图7C是根据本发明一个实施例的平滑图像的示意图；
图7D是根据本发明一个实施例的二值图像的示意图；
图7E是根据本发明一个实施例的对最终每点纹线距离的数据进行Gabor滤波后的结果示意图。
具体实施方式
下面详细描述本发明的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，旨在用于解释本发明，而不能理解为对本发明的限制。
为了解决相关技术中纹线距离算法存在的求得的纹线距离不准确、算法适用范围较窄等问题，本发明提出了一种纹线距离的计算方法和装置。下面参考附图描述本发明实施例的纹线距离的计算方法和装置。
图2是根据本发明一个实施例的纹线距离的计算方法的流程图。如图2所示，本发明实施例的纹线距离的计算方法，包括以下步骤：
S1，获取原始图像并进行灰度处理以生成灰度图像。
具体地，对原始图像并进行灰度处理，生成灰度图像A(i,j)。
S2，根据灰度图像生成法向图和切向图。
具体地，对灰度图像A(i,j)由梯度法可求出法向图O1(i,j)和切向图O2(i,j)。
S3，根据切向图对灰度图像进行滤波以生成平滑图像，并将平滑图像转换为二值图像。
在本发明的一个实施例中，根据切向图对灰度图像进行滤波以生成平滑图像具体包括：根据切向图对灰度图像进行1*7均值滤波以生成平滑图像。
具体地，利用切向图O2(i,j)对灰度图像A(i,j)进行1*7均值滤波去毛刺，得到平滑图像B(i,j)，然后利用差分二值化将平滑图像B(i,j)转变成二值图像D(i,j)。
S4，对二值图像进行分块，并根据法向图确定每个分块的中心点的法向方向。
具体地，把二值图像D(i,j)分成N*N(例如，N取33)大小的块，其中，块是逐点滑动的，所以，块跟块之间是有重叠的。
进一步地，在法向图O1(i,j)中读取每个分块的中心点的法向方向。
S5，在每个分块内对在每个分块的中心点的法向方向上的像素进行遍历，以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数以及变化处对应边界点的坐标和亚像素值，其中,第一像素值为脊线所在的像素的像素值，第二像素值为谷线所在的像素的像素值,也就是说,第一像素值的像素为脊线所在的像素，第二像素值的像素为谷线所在的像素。
在本发明的一个实施例中，第一像素值为0，第二像素值为1。下文均以此为例进行说明。
具体地，对于二值图像D(i,j)的每个分块，在每个分块内，对每个分块中心点的法向方向上的像素进行遍历，计算相邻两个像素的像素值由0变为1和由1变为0的变化次数以及变化处的像素坐标(即边界点的坐标)，如图3所示，条纹区域表示脊线，像素值为0，空白区域是谷线，像素值是1，箭头所指位置是由0变为1的位置，也就是边界点的位置，示意图3中由0到1变化的次数为3次。其中，当相邻两个像素的像素值由0变为1时，我们可以将边界点的坐标统一记为相邻两个像素中像素值为0的像素的坐标，也可以统一记为相邻两个像素中像素值为1的像素的坐标。
在本发明的一个实施例中，如图4所示，通过以下步骤生成变化处对应边界点的亚像素值：
S51，沿着预设方向获取与所述边界点相邻的两个像素的像素值。
S52，根据与所述边界点相邻的两个像素的像素值和所述边界点的像素值生成边界点的亚像素值。
具体地，在遍历的同时计算亚像素边界，分两种情况计算，一种是沿着倾斜方向计算边界点的亚像素，另一种是沿着竖直或者水平方向计算。下面分别就这两种情况进行说明。
在本发明的一个实施例中，当分块的中心点的法向方向所对应的法向角度等于0度时， 预设方向为竖直方向；当分块的中心点的法向方向所对应的法向角度等于90度时，预设方向为水平方向；且在分块的中心点的法向方向所对应的法向角度等于0度或90时，如果与边界点相邻的两个像素的像素值均与边界点的像素值相同，则边界点的亚像素值为0；如果与边界点相邻的两个像素的像素值中只有一个与边界点的像素值相同，则边界点的亚像素值为0.5；如果与边界点相邻的两个像素的像素值均与边界点的像素值不同，则边界点的亚像素值为1。
具体地，如图5A至图5C所示为沿着水平方向(如图5A中的y轴方向)计算亚像素取值的示意图，其中，在本发明中，以指纹图像(如图7A灰度图)的左上角为原点，指纹图像的竖直边界和水平边界分别为x轴和y轴建立坐标系，则所述分块的中心点的法向方向对应的法向角度为分块的中心点的法向方向与x轴的夹角，图中竖条纹填充块代表边界点，白色填充块代表谷线上的点、黑色填充块代表脊线上的点。在计算变化处对应边界点的亚像素值时，利用边界点两侧的两个像素的像素值进行判断，若都与边界点的像素值相同，则δ值(边界点的亚像素)为0；若有一个与边界点的像素值相同，δ取值为1/2；若都不与边界点的像素值相同时，δ取值为1。例如，图5A中与边界点相邻的两个像素均为白色填充块，图5B中与边界点相邻的两个像素分别为黑色和白色填充块，图5C中与边界点相邻的两个像素均为黑色填充块。
在本发明的另一个实施例中，当分块的中心点的法向方向所对应的法向角度不等于0度且不等于90度时，预设方向为除竖直方向和水平方向之外的方向(即倾斜方向)，如果与边界点相邻的两个像素的像素值均与边界点的像素值相同，则边界点的亚像素值为0；如果与边界点相邻的两个像素的像素值中只有一个与边界点的像素值相同，则边界点的亚像素值为0.25；如果与边界点相邻的两个像素的像素值均与边界点的像素值不同，则边界点的亚像素值为0.5。
具体地，如图6A至图6C所示为沿着非竖直或者水平方向(即倾斜方向，如图6A中所示)计算亚像素取值的示意图，其中，图中竖条纹填充块代表边界点，白色填充块代表谷线上的点、黑色填充块代表脊线上的点。在计算边界点的亚像素值时，利用其他两个相邻像素的像素值进行判断，若两个像素的像素值都与边界点的像素值相同，则δ值为0；有一个像素的像素值与边界点的像素值相同，δ值为1/4；若相邻两个像素的像素值都不与边界点的像素值相同，δ值为1/2。例如，图6A中与边界点相邻的两个像素均为白色填充块，图6B中与边界点相邻的两个像素分别为黑色和白色填充块，图6C中与边界点相邻的两个像素均为黑色填充块。
S6，根据每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数以及变化处对应边界点的坐标和亚像素值生成纹线距离,其中所述纹线距离为每个分 块内的中心点的纹线距离。
在本发明的一个实施例中，通过以下公式生成一个分块的中心点的纹线距离：
Figure PCTCN2016070192-appb-000001
Figure PCTCN2016070192-appb-000002
Figure PCTCN2016070192-appb-000003
其中，num1和num2为像素值在第一像素值和像素值为第二像素值之间的变化次数，num1为该分块内的相邻两个像素的像素值从第二像素值变化为第一像素值的次数，num2为为该分块内的相邻两个像素的像素值从第一像素值变化为第二像素值的次数，X1和Xnum1分别为该分块内沿该分块的中心点的法向方向第1次出现相邻两个像素的像素值从第二像素值变化为第一像素值处和第num1次出现相邻两个像素的像素值从第二像素值变化为第一像素值处对应的边界点的横坐标值，Y1和Ynum2分别为该分块内沿该分块的中心点的法向方向第1次出现相邻两个像素的像素值从第一像素值变化为第二像素值处和第num2次出现相邻两个像素的像素值从第一像素值变化为第二像素值处对应的边界点的横坐标值，θ为该分块的中心点的法向角度，取值范围是0到π，δXi为在该分块内沿该分块的中心点的法向方向第i次出现相邻两个像素的像素值从第二像素值变化为第一像素值处对应的边界点的亚像素值，δYi为在该分块内沿该分块的中心点的法向方向第i次出现相邻两个像素的像素值从第一像素值变化为第二像素值处对应的边界点的亚像素值，D1(i,j)和D2(i,j)分别为根据相邻两个像素的像素值从第二像素值变化为第一像素值和根据相邻两个像素的像素值从第一像素值变化为第二像素值对应计算出的距离，D(i,j)为该分块的中心点的纹线距离。
如此，即可计算得出每个分块的中心点的纹线距离。
在本发明的一个实施例中，还包括：根据每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数获取每个分块内的边界点的个数；对于分块内的边界点的个数小于预设个数的分块，则进一步在该分块内对该分块的中心点的法向方向的反方向上的像素进行遍历。
具体地，此处需要说明的是，上述D1(i,j)和D2(i,j)需要满足的条件为：①D1、D2 在每个分块中必须同时具有大于等于2个的由像素值0变为像素值1或由像素值1变为像素值0值转换的边界点，即：一个纹线距离的运算需要两条脊线和一条谷线来完成，或者由两个谷线和一条脊线来完成。如果不够，则D1或D2不存在；②如果D1或D2有一个不存在，则另一个D1或D2需要在法向方向相反的方向上寻找至少2个转换的边界点。
本发明实施例的纹线距离的计算方法，通过寻找指纹脊线和谷线的边界点，并根据边界点的坐标和亚像素值计算纹线距离，使得纹线距离更加准确，更接近指纹真实特征，从而更准确的反映了指纹全局疏密特征，而且该方法的抗噪声能力强，适用范围更广。
在本发明的一个实施例中，在S6之后，还包括：对纹线距离进行5*5局部区域均值滤波。
具体地，对计算出的纹线距离进行5*5局部区域均值滤波，以起到平滑的作用，得到最终每点纹线距离。
另外，为了使本发明实施例的纹线距离的计算方法的每个步骤的计算结果更加直观，给出该方法每步的效果图，如图7A所示为灰度图像的示意图，图7B所示为切向图的示意图，图7C所示为平滑图像的示意图，图7D所示为二值图像的示意图，图7E所示为对最终每点纹线距离的数据进行Gabor滤波后的结果示意图。
本发明实施例的纹线距离的计算方法，避免了相关技术中求取指纹正弦曲线极值比较复杂的情况，如：理论上有一个极大值点的地方，出现了大于2个的不确定数量的极大值点，不能准确求出指纹纹线距离；本发明实施例的方法是找指纹脊线和谷线的边界点，确定只有一个点，不会出现不确定数量的边界点，故对含有噪声图片冗余度高，条件要求不苛刻,扩大了适用范围。该方法具有很高的工程应用价值，可以为后期的图像滤波、分割、脊线跟踪和匹配提供可靠参数。
与上述几种实施例提供的纹线距离的计算方法相对应，本发明的一种实施例还提供一种纹线距离的计算装置，由于本发明实施例提供的纹线距离的计算装置与上述几种实施例提供的纹线距离的计算方法相对应，因此在前述纹线距离的计算方法实施方式也适用于本实施例提供的纹线距离的计算方法装置，在本实施例中不再详细描述。图8是根据本发明一个实施例的纹线距离的计算装置的结构示意图。如图8所示，本发明实施例的纹线距离的计算装置，包括：灰度处理模块100、生成模块200、平滑处理模块300、分块处理模块400、亚像素计算模块500和纹线距离生成模块600。
其中，灰度处理模块100用于获取原始图像并进行灰度处理以生成灰度图像。
生成模块200用于根据灰度图像生成法向图和切向图。
平滑处理模块300用于根据切向图对灰度图像进行滤波以生成平滑图像，并将平滑图像转换为二值图像。
在本发明的一个实施例中，平滑处理模块300具体用于：根据切向图对灰度图像进行1*7均值滤波以生成平滑图像，并将平滑图像转换为二值图像。
分块处理模块400用于对二值图像进行分块，并根据法向图确定每个分块的中心点的法向方向。
亚像素计算模块500用于在每个分块内对在每个分块的中心点的法向方向上的像素进行遍历，以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数、变化时对应边界点的坐标和亚像素值，其中，第一像素值为脊线所在的像素的像素值，第二像素值为谷线所在的像素的像素值。
在本发明的一个实施例中，第一像素值为0，第二像素值为1。
在本发明的一个实施例中，则亚像素计算模块500还用于：根据每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数获取每个分块内的边界点的个数，并对于分块内的边界点的个数小于预设个数的分块，在该分块内对该分块的中心点的法向方向的反方向上的像素进行遍历。
在本发明的一个实施例中，亚像素计算模块500生成边界点的亚像素值，具体为：沿着预设方向获取与边界点相邻的两个像素的像素值，并根据与边界点相邻的两个像素的像素值和边界点的像素值计算边界点的亚像素值。
在本发明的一个实施例中，亚像素计算模块500在分块的中心点的法向方向对应的法向角度等于0度时，确定预设方向为竖直方向；在分块的中心点的法向方向所对应的法向角度等于90度时，确定预设方向为水平方向；并进一步用于在与边界点相邻的两个像素的像素值均与边界点的像素值相同时，生成边界点的亚像素值为0；在与边界点相邻的两个像素的像素值中只有一个与边界点的像素值相同时，生成边界点的亚像素值为0.5；在与边界点相邻的两个像素的像素值均与边界点的像素值不同时，生成边界点的亚像素值为1。
在本发明的另一个实施例中，亚像素计算模块500在分块的中心点的法向方向对应的法向角度不等于0度且不等于90度时，确定预设方向为除竖直方向和水平方向之外的方向；并进一步用于在与边界点相邻的两个像素的像素值均与边界点的像素值相同时，生成边界点的亚像素值为0；在与边界点相邻的两个像素的像素值中只有一个与边界点的像素值相同时，生成边界点的亚像素值为0.25；在与边界点相邻的两个像素的像素值均与边界点的像素值不同时，生成边界点的亚像素值为0.5。
纹线距离生成模块600用于根据每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数和变化处对应边界点的坐标和亚像素值生成纹线距离。
在本发明的一个实施例中，纹线距离生成模块600通过公式(1)、(2)和(3)生成纹线距离。
在本发明的一个实施例中，纹线距离生成模块600还用于：对纹线距离进行5*5局部区域均值滤波。
本发明实施例的纹线距离的计算装置，通过寻找指纹脊线和谷线的边界点，并根据边界点的坐标和亚像素值计算纹线距离，使得纹线距离更加准确，更接近指纹真实特征，从而更准确的反映了指纹全局疏密特征，而且抗噪声能力强，适用范围更广。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本发明的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本发明的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本发明的实施例所属技术领域的技术人员所理解。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，"计算机可读介质"可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解 译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本发明的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，该程序在执行时，包括方法实施例的步骤之一或其组合。
此外，在本发明各个实施例中的各功能单元可以集成在一个处理模块中，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模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。
上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本发明的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本发明的限制，本领域的普通技术人员在本发明的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (17)

  1. 一种纹线距离的计算方法，其特征在于，包括以下步骤：
    获取原始图像并进行灰度处理以生成灰度图像；
    根据所述灰度图像生成法向图和切向图；
    根据所述切向图对所述灰度图像进行滤波以生成平滑图像，并将所述平滑图像转换为二值图像；
    对所述二值图像进行分块，并根据所述法向图确定每个分块的中心点的法向方向；
    在所述每个分块内对在所述每个分块的中心点的法向方向上的像素进行遍历，以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数以及变化处对应边界点的坐标和亚像素值，其中，所述第一像素值为脊线所在的像素的像素值，所述第二像素值为谷线所在的像素的像素值；以及
    根据所述每个分块内的所述相邻两个像素的像素值为在第一像素值和第二像素值之间变化的次数以及所述变化处对应边界点的坐标和亚像素值，生成纹线距离。
  2. 如权利要求1所述的纹线距离的计算方法，其特征在于，还包括：
    根据所述每个分块内的所述相邻两个像素的像素值为在第一像素值和第二像素值之间变化的次数获取所述每个分块内的所述边界点的个数；
    对于分块内的所述边界点的个数小于预设个数的分块，则进一步在所述分块内对所述分块的中心点的法向方向的反方向上的像素进行遍历。
  3. 如权利要求1或2所述的纹线距离的计算方法，其特征在于，所述生成所述变化处对应边界点的亚像素值,具体为:
    沿着预设方向获取与所述边界点相邻的两个像素的像素值；
    根据与所述边界点相邻的两个像素的像素值和所述边界点的像素值生成所述边界点的亚像素值。
  4. 如权利要求3所述的纹线距离的计算方法，其特征在于，所述根据与所述边界点相邻的两个像素的像素值和所述边界点的像素值生成所述边界点的亚像素值，具体为：
    当所述分块的中心点的法向方向对应的法向角度等于0度时，确定所述预设方向为竖直方向；且
    在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值相同时，生成所述边界点的亚像素值为0；在与所述边界点相邻的两个像素的像素值中只有一个与所述边界点的像素值相同时，生成所述边界点的亚像素值为0.5；在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值不同时，生成所述边界点的亚像素值为1。
  5. 如权利要求3所述的纹线距离的计算方法，其特征在于，所述根据与所述边界点相邻的两个像素的像素值和所述边界点的像素值生成所述边界点的亚像素值，具体为：
    当所述分块的中心点的法向方向对应的法向角度等于90度时，确定所述预设方向为水平方向；且
    在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值相同时，生成所述边界点的亚像素值为0；在与所述边界点相邻的两个像素的像素值中只有一个与所述边界点的像素值相同时，生成所述边界点的亚像素值为0.5；在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值不同时，生成所述边界点的亚像素值为1。
  6. 如权利要求3所述的纹线距离的计算方法，其特征在于，所述根据与所述边界点相邻的两个像素的像素值和所述边界点的像素值生成所述边界点的亚像素值，具体为：
    当所述分块的中心点的法向方向对应的法向角度不等于0度且不等于90度时，确定所述预设方向为除竖直方向和水平方向之外的方向；且
    在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值相同时，生成所述边界点的亚像素值为0；在与所述边界点相邻的两个像素的像素值中只有一个与所述边界点的像素值相同时，生成所述边界点的亚像素值为0.25；在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值不同时，生成所述边界点的亚像素值为0.5。
  7. 如权利要求1-6任一项所述的纹线距离的计算方法，其特征在于，通过以下公式生成一个分块的中心点的纹线距离：
    Figure PCTCN2016070192-appb-100001
    Figure PCTCN2016070192-appb-100002
    Figure PCTCN2016070192-appb-100003
    其中，num1为所述分块内的相邻两个像素的像素值从所述第二像素值变化为所述第一像素值的次数，num2为所述分块内的相邻两个像素的像素值从所述第一像素值变化为所述第二像素值的次数，X1和Xnum1分别为在所述分块内沿所述分块的中心点的法向方向第1次出现相邻两个像素的像素值从所述第二像素值变化为所述第一像素值处和第num1次出现相邻两个像素的像素值从所述第二像素值变化为所述第一像素值处对应的边界点的横坐 标值，Y1和Ynum2分别为在所述分块内沿所述分块的中心点的法向方向第1次出现相邻两个像素的像素值从所述第一像素值变化为所述第二像素值处和第num2次出现相邻两个像素的像素值从所述第一像素值变化为所述第二像素值处对应的边界点的横坐标值，θ为所述分块的中心点的法向角度，取值范围是0到π，δXi为在所述分块内沿所述分块的中心点的法向方向第i次出现相邻两个像素的像素值从所述第二像素值变化为所述第一像素值处对应的边界点的亚像素值，δYi为在所述分块内沿所述分块的中心点的法向方向第i次出现相邻两个像素的像素值从所述第一像素值变化为所述第二像素值处对应的边界点的亚像素值，D1(i,j)和D2(i,j)分别为根据相邻两个像素的像素值从所述第二像素值变化为所述第一像素值和根据相邻两个像素的像素值从所述第一像素值变化为所述第二像素值对应计算出的距离，D(i,j)为所述分块的中心点的纹线距离。
  8. 如权利要求1-7任一项所述的纹线距离的计算方法，其特征在于，在所述根据所述每个分块内的所述相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数以及变化处对应边界点的坐标和亚像素值，生成纹线距离之后，还包括：
    对所述纹线距离进行5*5局部区域均值滤波。
  9. 一种纹线距离的计算装置，其特征在于，包括：
    灰度处理模块，用于获取原始图像并进行灰度处理以生成灰度图像；
    生成模块，用于根据所述灰度图像生成法向图和切向图；
    平滑处理模块，用于根据所述切向图对所述灰度图像进行滤波以生成平滑图像，并将所述平滑图像转换为二值图像；
    分块处理模块，用于对所述二值图像进行分块，并根据所述法向图确定每个分块的中心点的法向方向；
    亚像素计算模块，用于在所述每个分块内对在所述每个分块的中心点的法向方向上的像素进行遍历，以计算每个分块内的相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数、变化处对应边界点的坐标和亚像素值，其中，所述第一像素值为脊线所在的像素的像素值，所述第二像素值为谷线所在的像素的像素值；以及
    纹线距离生成模块，用于根据所述每个分块内的所述相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数和所述变化处对应边界点的坐标和亚像素值生成纹线距离。
  10. 如权利要求9所述的纹线距离的计算装置，其特征在于，所述亚像素计算模块，还用于：根据所述每个分块内的所述相邻两个像素的像素值在第一像素值和第二像素值之间变化的次数获取所述每个分块内的所述边界点的个数，并对于分块内的所述边界点的个数小于预设个数的分块，在所述分块内对所述分块的中心点的法向方向的反方向上的像素 进行遍历。
  11. 如权利要求9或10所述的纹线距离的计算装置，其特征在于，所述亚像素计算模块还用于：
    沿着预设方向获取与所述边界点相邻的两个像素的像素值，并根据与所述边界点相邻的两个像素的像素值所述边界点的像素值生成所述边界点的亚像素值。
  12. 如权利要求11所述的纹线距离的计算装置，其特征在于，所述亚像素计算模块还用于在所述分块的中心点的法向方向对应的法向角度等于0度时，确定所述预设方向为竖直方向；并进一步用于：
    在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值相同时，生成所述边界点的亚像素值为0；在与所述边界点相邻的两个像素的像素值中只有一个与所述边界点的像素值相同时，生成所述边界点的亚像素值为0.5；在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值不同时，生成所述边界点的亚像素值为1。
  13. 如权利要求11所述的纹线距离的计算装置，其特征在于，所述亚像素计算模块还用于在所述分块的中心点的法向方向对应的法向角度等于90度时，确定所述预设方向为水平方向；并进一步用于：
    在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值相同时，生成所述边界点的亚像素值为0；在与所述边界点相邻的两个像素的像素值中只有一个与所述边界点的像素值相同时，生成所述边界点的亚像素值为0.5；在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值不同时，生成所述边界点的亚像素值为1。
  14. 如权利要求11所述的纹线距离的计算装置，其特征在于，所述亚像素计算模块还用于在所述分块的中心点的法向方向对应的法向角度不等于0度且不等于90度时，确定所述预设方向为除竖直方向和水平方向之外的方向；并进一步用于：
    在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值相同时，生成所述边界点的亚像素值为0；在与所述边界点相邻的两个像素的像素值中只有一个与所述边界点的像素值相同时，生成所述边界点的亚像素值为0.25；在与所述边界点相邻的两个像素的像素值均与所述边界点的像素值不同时，生成所述边界点的亚像素值为0.5。
  15. 如权利要求9-14所述的纹线距离的计算装置，其特征在于，所述纹线距离生成模块还用与通过以下公式生成一个分块的中心点的纹线距离：
    Figure PCTCN2016070192-appb-100004
    Figure PCTCN2016070192-appb-100005
    Figure PCTCN2016070192-appb-100006
    其中，num1为所述分块内的相邻两个像素的像素值从所述第二像素值变化为所述第一像素值的次数，num2为所述分块内的相邻两个像素的像素值从所述第一像素值变化为所述第二像素值的次数，X1和Xnum1分别为在所述分块内沿所述分块的中心点的法向方向第1次出现相邻两个像素的像素值从所述第二像素值变化为所述第一像素值处和第num1次出现相邻两个像素的像素值从所述第二像素值变化为所述第一像素值处对应的边界点的横坐标值，Y1和Ynum2分别为在所述分块内沿所述分块的中心点的法向方向第1次出现相邻两个像素的像素值从所述第一像素值变化为所述第二像素值处和第num2次出现相邻两个像素的像素值从所述第一像素值变化为所述第二像素值处对应的边界点的横坐标值，θ为所述分块的中心点的法向角度，取值范围是0到π，δXi为在所述分块内沿所述分块的中心点的法向方向第i次出现相邻两个像素的像素值从所述第二像素值变化为所述第一像素值处对应的边界点的亚像素值，δYi为在所述分块内沿所述分块的中心点的法向方向第i次出现相邻两个像素的像素值从所述第一像素值变化为所述第二像素值处对应的边界点的亚像素值，D1(i,j)和D2(i,j)分别为根据相邻两个像素的像素值从所述第二像素值变化为所述第一像素值和根据相邻两个像素的像素值从所述第一像素值变化为所述第二像素值对应计算出的距离，D(i,j)为所述分块的中心点的纹线距离。
  16. 如权利要求9-15所述的纹线距离的计算装置，其特征在于，所述纹线距离生成模块，还用于：
    对所述纹线距离进行5*5局部区域均值滤波。
  17. 一种计算机可读存储介质，包括计算机指令，当所述计算机指令被执行时，使得执行根据权利要求1-8中任一项所述的纹线距离的计算方法。
PCT/CN2016/070192 2015-02-13 2016-01-05 纹线距离的计算方法和装置 WO2016127735A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2017542060A JP6494776B2 (ja) 2015-02-13 2016-01-05 線間距離計算方法、線間距離計算装置、コンピュータプログラム、及びコンピュータ読み取り可能な記憶媒体
EP16748536.6A EP3264361A4 (en) 2015-02-13 2016-01-05 Method and device for calculating ridge distance
KR1020177022541A KR101985689B1 (ko) 2015-02-13 2016-01-05 선거리 계산 방법 및 장치
US15/548,688 US20180018497A1 (en) 2015-02-13 2016-01-05 Method and device for calculating line distance

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510080690.0A CN105205802B (zh) 2015-02-13 2015-02-13 纹线距离的计算方法和装置
CN201510080690.0 2015-02-13

Publications (1)

Publication Number Publication Date
WO2016127735A1 true WO2016127735A1 (zh) 2016-08-18

Family

ID=54953464

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/070192 WO2016127735A1 (zh) 2015-02-13 2016-01-05 纹线距离的计算方法和装置

Country Status (7)

Country Link
US (1) US20180018497A1 (zh)
EP (1) EP3264361A4 (zh)
JP (1) JP6494776B2 (zh)
KR (1) KR101985689B1 (zh)
CN (1) CN105205802B (zh)
TW (1) TWI559235B (zh)
WO (1) WO2016127735A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110717502A (zh) * 2019-10-22 2020-01-21 大连大学 一种脉冲激光拼焊图像中熔池识别方法

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105205802B (zh) * 2015-02-13 2017-04-12 比亚迪股份有限公司 纹线距离的计算方法和装置
CN106503609B (zh) * 2015-09-08 2019-07-16 腾讯科技（深圳）有限公司 指纹纹线点的识别方法及装置
CN106529572B (zh) * 2016-10-14 2019-07-09 广州视源电子科技股份有限公司 色环电阻的检测方法、装置和自动光学检测系统
CN108121940A (zh) * 2016-11-29 2018-06-05 深圳指芯智能科技有限公司 一种指纹图像分析的方法和装置
CN109426795A (zh) * 2017-09-05 2019-03-05 比亚迪股份有限公司 指纹识别方法和装置
CN108038839B (zh) * 2017-11-23 2022-01-11 宁波大学 一种流水生产线上双绞线绞距实时检测方法
CN107967144B (zh) * 2017-12-08 2021-10-22 京东方科技集团股份有限公司 一种数据烧录方法和装置
CN111292344A (zh) * 2018-12-07 2020-06-16 致伸科技股份有限公司 摄像模块的检测方法
CN111161151A (zh) * 2019-12-30 2020-05-15 广东利元亨智能装备股份有限公司 图像拼接方法、装置、机器人和存储介质
CN112232159B (zh) * 2020-09-30 2021-12-07 墨奇科技（北京）有限公司 指纹识别的方法、装置、终端及存储介质
CN113284133B (zh) * 2021-06-16 2022-04-12 北京瑞莱智慧科技有限公司 一种光伏电池片间距的确定方法、装置、设备及存储介质
CN113483682B (zh) * 2021-06-16 2023-05-19 苏州赛腾精密电子股份有限公司 一种基于机器视觉的缝隙测量方法及系统
CN113421249B (zh) * 2021-06-30 2024-02-06 上海申瑞继保电气有限公司 变电站设备可变尺度图像数值处理方法
WO2023019132A1 (en) * 2021-08-10 2023-02-16 Deeplook, Inc. Radiomic systems and methods
CN113865488B (zh) * 2021-09-24 2023-10-27 北京京东方技术开发有限公司 一种距离测量方法、电子设备及计算机可读存储介质
CN116107394B (zh) * 2023-04-06 2023-08-04 合肥联宝信息技术有限公司 调整方法、装置、电子设备及存储介质
CN116958514B (zh) * 2023-09-20 2023-12-05 中国空气动力研究与发展中心高速空气动力研究所 一种光学图像激波位置的亚像素定位方法
CN117496560B (zh) * 2023-12-29 2024-04-19 深圳市魔力信息技术有限公司 一种基于多维向量的指纹纹线识别方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6961453B2 (en) * 2001-08-31 2005-11-01 Secugen Corporation Method for extracting fingerprint feature data using ridge orientation model
CN1737821A (zh) * 2005-08-15 2006-02-22 阜阳师范学院 自动指纹识别方法中的图像分割及指纹纹线距离提取技术
CN1818927A (zh) * 2006-03-23 2006-08-16 北京中控科技发展有限公司 指纹识别方法与系统
CN101271516A (zh) * 2008-04-02 2008-09-24 范九伦 指纹图像的方向滤波增强方法
CN102930241A (zh) * 2012-08-03 2013-02-13 北京天诚盛业科技有限公司 指纹图像的处理方法和处理装置
CN105205802A (zh) * 2015-02-13 2015-12-30 比亚迪股份有限公司 纹线距离的计算方法和装置

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR0126464B1 (ko) * 1991-06-03 1998-04-02 박원희 지문 인식 방법
JP2725599B2 (ja) * 1994-06-21 1998-03-11 日本電気株式会社 隆線方向抽出装置
US7142699B2 (en) * 2001-12-14 2006-11-28 Siemens Corporate Research, Inc. Fingerprint matching using ridge feature maps
JP4075458B2 (ja) * 2002-05-21 2008-04-16 松下電器産業株式会社 指紋照合装置
US7120280B2 (en) * 2002-09-27 2006-10-10 Symbol Technologies, Inc. Fingerprint template generation, verification and identification system
JP4270842B2 (ja) * 2002-10-22 2009-06-03 セコム株式会社 指紋照合装置
KR100825773B1 (ko) * 2005-08-23 2008-04-28 삼성전자주식회사 방향 추정 방법 및 장치
CN101329725B (zh) * 2008-07-30 2010-10-06 电子科技大学 基于梯度投影和形态学的指纹图像分割方法
TW201023055A (en) * 2008-12-12 2010-06-16 Moredna Technology Co Ltd Highly efficient method for processing fingerprint images
CN101576999B (zh) * 2009-06-01 2012-10-17 杭州晟元芯片技术有限公司 一种指纹图像方向增强方法
CN101782966B (zh) * 2010-02-11 2012-07-04 上海点佰趣信息科技有限公司 指纹图像二值化处理方法
US9070002B2 (en) * 2011-10-18 2015-06-30 King Fahd University Of Petroleum And Minerals Method of performing fingerprint matching

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6961453B2 (en) * 2001-08-31 2005-11-01 Secugen Corporation Method for extracting fingerprint feature data using ridge orientation model
CN1737821A (zh) * 2005-08-15 2006-02-22 阜阳师范学院 自动指纹识别方法中的图像分割及指纹纹线距离提取技术
CN1818927A (zh) * 2006-03-23 2006-08-16 北京中控科技发展有限公司 指纹识别方法与系统
CN101271516A (zh) * 2008-04-02 2008-09-24 范九伦 指纹图像的方向滤波增强方法
CN102930241A (zh) * 2012-08-03 2013-02-13 北京天诚盛业科技有限公司 指纹图像的处理方法和处理装置
CN105205802A (zh) * 2015-02-13 2015-12-30 比亚迪股份有限公司 纹线距离的计算方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3264361A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110717502A (zh) * 2019-10-22 2020-01-21 大连大学 一种脉冲激光拼焊图像中熔池识别方法
CN110717502B (zh) * 2019-10-22 2023-09-05 大连大学 一种脉冲激光拼焊图像中熔池识别方法

Also Published As

Publication number Publication date
KR101985689B1 (ko) 2019-09-03
JP6494776B2 (ja) 2019-04-03
CN105205802B (zh) 2017-04-12
KR20170105560A (ko) 2017-09-19
EP3264361A1 (en) 2018-01-03
CN105205802A (zh) 2015-12-30
TW201629850A (zh) 2016-08-16
EP3264361A4 (en) 2018-10-31
TWI559235B (zh) 2016-11-21
US20180018497A1 (en) 2018-01-18
JP2018508891A (ja) 2018-03-29

Similar Documents

Publication Publication Date Title
WO2016127735A1 (zh) 纹线距离的计算方法和装置
US9747498B2 (en) Biometric recognition
Spreeuwers Fast and accurate 3D face recognition: using registration to an intrinsic coordinate system and fusion of multiple region classifiers
Dibeklioglu et al. 3D facial landmarking under expression, pose, and occlusion variations
WO2016127736A1 (zh) 指纹重叠区域面积的计算方法及电子装置
CN100494886C (zh) 三维扫描系统中圆形标志点的检测方法
CN108830888B (zh) 基于改进的多尺度协方差矩阵特征描述子的粗匹配方法
US11023762B2 (en) Independently processing plurality of regions of interest
WO2017113736A1 (zh) 手指与手腕的区分方法及其装置
CN106936964A (zh) 一种基于霍夫变换模板匹配的手机屏幕角点检测方法
CN104392229A (zh) 基于笔画片段方向分布特征的笔迹识别方法
Maiwald et al. Feature matching of historical images based on geometry of quadrilaterals
CN102831578B (zh) 图像处理方法和图像处理设备
US11450140B2 (en) Independently processing plurality of regions of interest
Jahanbin et al. Passive three dimensional face recognition using iso-geodesic contours and procrustes analysis
KR20070088982A (ko) 변형에 강건한 홍채 인식 방법
WO2018032703A1 (zh) 跟踪指部轮廓的方法及其装置
US20220027649A1 (en) Method and apparatus with liveness detection
EP3827409B1 (en) Autocalibration for multiple cameras using near-infrared illuminators
CN106605235A (zh) 用于确定是否在扫描操作期间存在目标对象的方法及系统
WO2018032704A1 (zh) 跟踪指部轮廓的方法及其装置
Bajahzar Fingerprint Smoothing Using Different Interpolation Techniques
Jiang et al. Comprehensive Study on Shape Representation Methods for Shape-Based Object Recognition
Li et al. 3D face recognition for security and defense engineering
Medina-Melendrez et al. Irregular Ellipse Detection Method by Confined Space Sweep

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: 16748536

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15548688

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2017542060

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20177022541

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2016748536

Country of ref document: EP