JP5049074B2 - Apparatus and method for determining acceptability of analyzed fingerprint image - Google Patents

Description

  The present invention relates to an apparatus and method for determining the acceptability of an analyzed fingerprint image.

  As the capabilities of mobile devices increase, more personal information is stored on the mobile devices. This trend has increased the demand for more stringent safety measures than commonly adopted safety measures, such as personal identification numbers (PINs).

  One proposed solution is to provide a sensor for fingerprint confirmation. Such additional sensors increase costs. Therefore, it is desirable to utilize components that exist within existing mobile devices. However, the adoption of a built-in camera leaves a fundamental problem in determining an acceptable image for fingerprint confirmation.

  Korean Patent Application No. 2006-0078980 filed with the Korean Patent Office on August 21, 2006, entitled “Apparatus and Method for Determining Accommodability of Analyzed Fingerprint Image”, and October 17, 2006 US Provisional Application No. 60 / 852,014, filed with the US Patent and Trademark Office on the day, was referenced in this application.

US Pat. No. 6,289,112 US Pat. No. 6,763,127 US Pat. No. 7,073,711 US Pat. No. 7,072,523 US Pat. No. 6,314,197 The Fourth IEEE Workshop on Automatic Identification Advanced Technologies, October 17-18, 2005

It is an object of the present invention to provide an apparatus and method for estimating the focus of a fingerprint image by dividing the image into blocks and utilizing repetitive patterns of the fingerprint image, ie, fingerprint ridges and valleys. is there.
Another object of the present invention is to provide an apparatus and method for estimating a quality index of a fingerprint image by removing an area where the image is not well focused and an area other than the finger.
It is still another object of the present invention to provide an apparatus and method for estimating the tilt and undulation of an imaged finger.

The method and apparatus for determining the acceptability of an analyzed fingerprint image according to the present invention substantially overcomes a number of problems due to limitations and weaknesses of the prior art.
At least one of the objectives is achieved by providing a method for determining acceptability of a fingerprint image to be analyzed. A method for determining acceptability of a fingerprint image to be analyzed includes dividing an image into a plurality of blocks, determining a focus for each block, and enabling the block if the focus is sufficiently well aligned. Determining the validity of a block, otherwise determining the block as an invalid block, summing the number of valid blocks in the image, and the image if the number of valid blocks is sufficient Determining the validity of the image.

If the ratio of the number of effective blocks to the number of blocks exceeds a predetermined critical value, the number of effective blocks may be sufficient.
Determining the effectiveness of the block may include comparing the measured focus with a focus critical value determined from the blur model.

  The method of determining acceptability of the analyzed fingerprint image further comprises, for each block, determining whether the block contains a repetitive pattern, wherein the validity of the block is determined. The step includes determining the block as a valid block if the block is well focused and has a repetitive pattern, otherwise the block is determined as an invalid block. Determining whether a block includes a repetitive pattern may include determining a quality index that includes the consistency and symmetry of the block's gradient distribution. The quality index may be the sum of the consistency and symmetry weights.

  A method for determining acceptability of a fingerprint image to be analyzed includes detecting a center point and a contour line of the image, and determining undulations and slopes at the center point and contour line if the image is valid. And, when the undulation and slope are sufficiently small, selecting an image for subsequent processing.

  The steps of determining the undulation degree include determining a left side distance between the center point and the left boundary of the contour line, determining a right side distance between the center point and the right boundary of the contour line, and a right side distance. Determining an undulation proportional to the ratio between the difference between the absolute values of the left and right distances and the sum of the right and left distances.

  Determining the slope can include determining an upward slope and determining a downward slope. The step of determining the slope includes a step of determining a first distance from the center point to the upper boundary of the effective area, a step of determining a second distance from the center point to the upper part of the contour line, and an effective area from the center point. Determining a third distance to a lower boundary of the first, wherein the upward slope is proportional to the ratio of the first distance to the second distance, and the downward slope is a ratio of the third distance to the second distance. Is proportional to The upper boundary and the lower boundary of the effective area can be determined such that the width of the effective area is larger than half the width of the left boundary and the right boundary of the contour line in both the upper boundary and the lower boundary.

  Determining the slope includes determining a left distance DL between the center point and the left boundary of the contour; determining a right distance DR between the center point and the right boundary of the contour; The step of determining the upper distance DF from the center point to the upper boundary of the effective area, the step of determining the lower distance DB from the center point to the lower boundary of the effective area, and a ratio between the width and height of the effective area Determining N, where the slope is N− (DF + DB) / (DL + DR).

The upper boundary and the lower boundary of the effective area are determined so that the width of the effective area is larger than half the width of the left boundary and the right boundary of the contour line.
The step of generating an image may be realized using a camera, for example a camera that is part of a portable device.

  Determining the validity of a block may include utilizing a VMLOG (Variance-Modified-Lapalasian of Gaussian) method. The VMLOG method may include at least one of the following equations.

  At least one of the objectives is achieved by providing a method for determining acceptability of a fingerprint image to be analyzed. The method of determining acceptability of the analyzed fingerprint image includes dividing the image into a plurality of blocks, and for each block, determining whether the block includes a repetitive pattern; Determining the validity of a block that determines the block as a valid block if the block includes a repetitive pattern, and otherwise determining the block as an invalid block, and summing the number of valid blocks in the image And determining that the image is valid if the number of valid blocks is acceptable.

  If the ratio of the number of effective blocks to the number of multiple blocks exceeds a predetermined critical value, the number of effective blocks may be acceptable.

  At least one of the objects provides a method for determining acceptability of a fingerprint image in a product having a machine accessible medium that includes data that causes the machine to determine the acceptability of a fingerprint image to be analyzed. Is realized. A method for determining the acceptability of a fingerprint image includes the steps of dividing the image into a plurality of blocks, determining for each block whether or not the block is well focused, and focusing the block well. If so, determine the state of the block, determining the block as a valid block, otherwise determining the block as an invalid block, summing the number of valid blocks in the image, and accepting the number of valid blocks. Determining the image as valid. If the ratio of the number of effective blocks to the number of multiple blocks exceeds a predetermined critical value, the number of effective blocks may be acceptable.

  At least one of the objects is achieved by providing an apparatus for determining acceptability of a fingerprint image to be analyzed. The apparatus includes: a divider that divides an image into a plurality of blocks; a focus estimator that determines whether the block is sufficiently focused for each block; a block focus and a focus critical value; In comparison, if the block is sufficiently focused, the block is determined as a valid block, otherwise the block is determined as an invalid block, and the number of valid blocks in the image is summed. A validity determiner that determines the image as valid if the number of valid blocks is acceptable.

  At least one of the objectives is achieved by providing a method for determining acceptability of a fingerprint image to be analyzed. The method includes detecting a center point and a contour of an image, determining a undulation and slope from the center and contour, and when the undulation and slope are acceptable, Selecting an image for the purpose.

  Determining the undulation degree includes determining a left-side distance between a center point and a left-side boundary of the contour line; determining a right-side distance between the center point and the right-side boundary of the contour line; Determining an undulation proportional to a ratio between an absolute value difference between the left distance and the right distance and a sum of the left distance and the right distance.

The method further includes determining an effective area of the fingerprint image, where the effective area includes a repetitive pattern.
The step of determining the slope may include a step of determining an upward slope and a step of determining a downward slope.

The method includes determining a first distance from the center point to the upper boundary of the effective region, determining a second distance from the center point to the upper portion of the contour line, and from the center point to the lower boundary of the effective region. Determining a third distance of the second distance, wherein the upward slope is proportional to the ratio of the first distance to the second distance, and the downward slope is proportional to the ratio of the third distance to the second distance.
The upper and lower boundaries of the effective area are determined such that the width of the effective area is larger than half the width of the left and right boundaries of the contour line in both the upper and lower boundaries.

  Determining the slope includes determining a left distance DL between the center point and the left boundary of the contour; determining a right distance DR between the center point and the right boundary of the contour; Determining the upper distance DF from the center point to the upper boundary of the effective area; determining the lower distance DB from the center point to the lower boundary of the effective area; the upper distance and the lower distance versus the left distance and the right side Determining a slope according to a ratio of distances. The upper boundary and the lower boundary of the effective area are determined such that the width of the effective area is larger than half of the width of the left boundary and the right boundary of the outline in both the upper boundary and the lower boundary.

  At least one of the objectives provides a method for determining acceptability of a fingerprint image in a product having a machine accessible medium that includes data that causes the machine to determine acceptability of a fingerprint image to be analyzed. It is realized by doing. A method for determining acceptability of a fingerprint image includes steps of detecting a center point and an outline of an image, determining an undulation and inclination from the center point and the outline, and the undulation and inclination are sufficient. The image may be selected for subsequent processing.

  At least one of the objectives is achieved by providing a system for determining acceptability of a fingerprint image to be analyzed. The system determines a center point and contour of the image, and determines the undulations and slopes from the center points and contours, and when the undulations and slopes are small enough for subsequent processing And a processor for selecting the image.

  According to the present invention, the repetitive pattern of a fingerprint image can be used to estimate the focus of a finger, and the quality can be estimated by removing an out-of-focus area and an image area other than the finger, and the image is imaged. The degree of finger tilt and undulation can be estimated. Therefore, an image suitable for fingerprint recognition can be generated.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the invention may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. The same reference numerals provided in each drawing represent the same member.
A method and apparatus for determining acceptability of a fingerprint image for fingerprint verification is described in detail below. First, as shown in FIG. 1, basic image formation will be outlined.

  Generally, the best focused image is the best quality image. The autofocus function is commonly used to achieve such a well-focused image. The autofocus function mainly includes calculating the focus for the image at different lens positions and moving the lens to the maximum focus position. The relationship between the object distance u, the focal distance f of the lens, and the image distance v is explained by the law of Gaussian lenses. The law of the Gaussian lens is expressed by the following equation (1).

  When the object point P is accurately focused, the image distance v is the same as the sensor sensing distance s, and the image plane corresponds to the sensor plane. If the sensor surface moves an sv distance from the image surface, the object point P is blurred over a circular region having a radius R (σb) on the sensor surface. The resulting light energy distribution in the circular region is approximated by the following equation (2).

  A focused image contains more high frequency information than an out-of-focus image, and most autofocus techniques rely on such differences. However, the use of the autofocus function for acquiring a fingerprint image has a problem as described later, and generally a special tool for assuring an appropriate position of the fingerprint is required. However, additional tools are not an optimal solution for users and producers.

Even if the appropriate distance between the finger and the sensor is determined, other factors still affect the fingerprint image processing. For example, when an object is focused close to the sensor surface, the shape of the object affects the degree of focus. The area perpendicular to the optical axis of the camera is well focused.
However, since the finger is cylindrical and the fingertips become progressively thinner, many fingerprint images are considered out of focus. If only the well-focused part of the fingerprint image is considered valid, there will be insufficient images available for fingerprint recognition. Moreover, because of the bent shape of the finger, an image in which the undulation and inclination of the finger are not suitable for fingerprint recognition is generated.

  In a preferred embodiment of the present invention, a system and method for determining the acceptability of a fingerprint image for fingerprint recognition may comprise three components to address at least the aforementioned issues.

  First, focus estimation can estimate the focus of a finger by dividing the image into blocks and by utilizing a repetitive pattern of the fingerprint image, such as a ridge and valley pattern. Second, the quality estimation can estimate the quality index of the effective area by removing the out-of-focus area and the image area other than the finger. These two components are disclosed in “Real-time image selection algorithm: fingerprint recognition of a mobile device having a built-in camera” (Non-Patent Document 1), which is a paper by many outside Lee Dong-jae. As integrated. Third, pose estimation can estimate the degree of tilt and undulation of the imaged fingers. An acceptable image is selected for fingerprint recognition based on at least one of the three components.

  FIG. 2 is a flowchart illustrating a process according to an exemplary embodiment of the present invention. First, in step 202, for example, an image from a camera, scanner, or other image input device is received as an input image. In step 204, the image is divided into blocks. At step 206, the focus is calculated. At step 208, a quality index is calculated. At step 210, a first determination is made regarding the validity of the image for fingerprint analysis.

  If it is determined at step 210 that the image is invalid, another image is input and the process begins again at step 202. If step 210 determines that the image is valid, the process proceeds to step 222.

  At step 222, the finger pose to be imaged is estimated, for example, by detecting the contour and center point of the image. At step 224, the imaged finger undulation and tilt are estimated. At step 230, a second determination is made regarding the acceptability of the image for fingerprint analysis. If step 230 determines that the image is not acceptable, another image is input and the process begins again at step 202. If step 230 determines that the image is acceptable, then step 242 selects the image.

  FIG. 3 is a block diagram of a fingerprint analyzer 350 for analyzing an acceptable fingerprint image and a system for determining acceptability according to a preferred embodiment of the present invention. The system includes an image input 302, a splitter 304, a focus estimator 306, a focus determiner 308, a quality index (QI) estimator 310, a quality index determiner 312, a validity determiner 314, a fingerprint pose analyzer 322, a relief. An estimator 324, a slope estimator 326, a undulation determiner 328, a slope determiner 330, and a final determiner 340 are provided. The function of the block can be realized as a traditional circuit well known to those skilled in the art. Hereinafter, the block corresponding to the step is inserted in parentheses.

  First, the image received in step 202 (image input 302) is divided into m * m blocks in step 204 (divider 304), and the degree of focus is estimated for each block. As previously mentioned, conventional methods are sufficient for images with a lot of high frequency information, but are not suitable for close-up and repetitive images. According to an embodiment of the present invention, the focus measurement of step 206 (focus estimator 306) is determined using VMLOG. The changed LOG is given by the following equation (3).

Here, I is an input image, G is a Gaussian filter, and * is a convolution operator.
In order to accommodate possible changes in the size of the repetitive pattern, inter-pixel spacing or step variables are utilized. The discrete approximation of the changed LOG is given by the following equations (4) to (6).

Here, ω0 is a positive value, and the remaining kernel is a negative value. The kernel and step values are determined by repetitive pattern spacing.
Finally, the variance of MLOG is calculated as follows:

  Here, VAR is a dispersion operator. The degree of focus for each such block is compared to a critical value that determines whether the block is well focused.

  The critical value with which VMLOG is compared is determined using a blur model. For long range images, the degree of blur R is similar throughout the image and the focus measurement is calculated over the entire image. On the other hand, in a short-range image used for fingerprint analysis, the degree of blur between the object region and the background is very different. Such a difference is used to distinguish the fingerprint region and the background region of the image.

  FIG. 4 shows the relationship between the degree of blur R and the distance u from the lens to the object. When the image is in focus, u0 is the distance from the object to the lens, and v0 is the distance between the lens and the image plane. When the image is out of focus, each such distance is represented by the u and v variables. The blur effect is modeled as follows.

Although the sufficiency of focus is evaluated by selecting an appropriate blur critical value Tb, the appropriate blur critical value Tb is used to determine the minimum focus critical value Tf.
In addition to calculating the degree of focus, the quality index is calculated in step 208 (QI estimator 310). The quality index calculation can determine whether the image has a pattern that matches the pattern of a properly imaged fingerprint, i.e., has a linear repeating pattern.

  If the input image has an ideal linear pattern as shown in FIG. 5A, the image gradient is consistent and symmetrical in the direction perpendicular to the fingerprint ridge direction, as shown in FIG. 5B. Distributed. In a typical image as shown in FIG. 5C, the gradients are dense as shown in FIG. 5D. If the finger is too close as shown in FIG. 5E, the gradient is consistent but not symmetric as shown in FIG. 5F. Therefore, these two characteristics, consistency and symmetry, are investigated to determine image quality.

  The consistency is calculated using the following equation (9).

Here, Gx is an x-axis gradient, and Gy is a y-axis gradient.
The degree of symmetry is a ratio between the number of gradients on the direction line and the number of gradients below the direction line in the gradient distribution, and is given by the following equation (10).

  Here, Np is the number of gradients on the direction line in FIG. 5B, FIG. 5D or FIG. 5F, and Nn is the number of gradients on the direction line in FIG. 5B, FIG. 5D or FIG.

  The quality index (QI) is a combination of these two attributes, and is expressed by the following equation (11).

  Here, ω1 and ω2 are weight values that vary depending on the characteristics of the input device being used.

  The validity of each block is determined as shown in FIG. Both VMLOG and QI are each compared to an experimentally established critical value, which determines whether the block is valid. The criticality value Tf of the degree of focus is obtained through an average of the number of blur models or other statistical analysis. The quality index critical value TQI is determined by selecting several images from the entire test image and by training using statistical analysis, eg, least mean squares.

  In the special configuration shown in FIG. 6, if VMLOG is greater than Tf at step 212 (focus determiner 308), flow proceeds to step 208 (QI estimator 310). Otherwise, the block is determined to be invalid and the flow returns to step 206 (focus estimator 306) to determine the validity of the next block. If the QI is greater than the TQI in step 214 (QI estimator 310), the block is determined to be valid. Otherwise, the block is determined to be invalid and the flow returns to step 206 (focus estimator 306) to determine the validity of the next block.

  Obviously, the order of comparison may be reversed or performed simultaneously. The use of a block for evaluating a fingerprint image including directionality is disclosed in, for example, Patent Document 1 entitled “System and Method for Determining Block Direction in Fingerprint Images”.

  To determine whether the flow proceeds to step 222 (fingerprint pose analyzer 322) of FIG. 2, RVB, which is the ratio of the number of valid blocks NV to the total number of blocks NT, is obtained at step 216 of FIG. As shown in the sex determiner 314), it is compared with the critical value TR. If RVB is greater than TR, go to step 222 (fingerprint pose analyzer 322), otherwise the image is discarded and the flow returns to step 202 (image input 302). Therefore, the first determination step of step 210 can be performed as shown in FIGS.

  Even if the finger is properly focused, the finger pose is changed due to the undulation and tilt of the finger, especially when no additional device for fixing the finger is employed. A sudden pose change between images reduces the general area and reduces matching performance. Thus, the pose change is evaluated to determine whether the image is recognizable. According to the embodiment of the present invention, the pose change is estimated using the center point and the outline of the finger.

  In addition to the finger outline and center point, the effective area is formed from effective blocks. In particular, the morphological operator is used to remove small effective areas, otherwise it is used to smooth the image. This is disclosed in, for example, Patent Document 2 entitled “Apparatus and Method for Fingerprint Recognition System” and was referred to in this application. If there are one or more effective areas, the largest effective area is selected. If the invalid area exists in the largest effective area, the invalid area is absorbed by the effective area. The relationship among the contour line, center point, effective area and finger position is shown in FIG.

  The outline of the finger is extracted using the difference between a series of effective images. For example, it is extracted using two binary subtraction images as shown in FIG. 8A shows an nth frame image, FIG. 8B shows an n-1th frame image, FIG. 8C shows a SIP image, and FIG. 8D shows a SIN image. FIG. 8E shows a DI image, and FIG. 8F shows a foreground image. Such a relationship is expressed by the following equations (12) to (14).

  The methodology described above employs two images. Even if more data points need to be analyzed and correlated in a single image and the contour extraction takes more time, the contour extraction is realized using a single image. The

Several algorithms are known to detect only one point in the fingerprint image and to estimate the upper center point for point estimation, but a simple and well-known algorithm for determining the approximate position, eg Poincare index Is used. At the center point, the Poincare index is ½. The center point can be detected only on the effective block.
Once the contour, center point, and effective area are detected, the finger pose is analyzed as described below.

  A finger undulation is a rotation of a finger to the right or left side along the longitudinal axis of the finger. The undulation degree R0 is determined using the distance R1 from the center point to the left side surface of the finger contour line and the distance R2 from the center point to the right side surface of the finger contour line, and is given by the following equation (15).

  As shown in FIG. 5, the degree of undulation becomes closer to 0 as the position of the center point approaches the center of the finger. The undulation level becomes closer to 1 as the center point goes to both sides.

  The tilt of the finger is in the direction above or below the finger. If the finger is inclined, the surface area of the finger orthogonal to the image plane is reduced. If the depth of the camera image plane is fixed, the increase in the tilt ratio reduces the area of the effective area. Even when the slope ratio is small enough to obtain a sufficiently large effective area, the image is still unrecognizable. For example, if the center point is located near the boundary of the effective area or is not within the effective area, the image is not selected.

  As shown in FIG. 9, the boundary lines LU and LB are determined at the minimum point where the width of the effective area is wider than half the finger width. The distance P1 is a distance from the center point to the upper boundary line LU, and the distance P3 is a distance from the center point to the lower boundary line LB. The inclination ratio Plate is approximated by the following equation (16).

  Here, N is a number larger than the ratio between the width of the finger and the length of the effective area in the vertical direction. The PPOS that is the center point position in the effective region is estimated by the following equation (17).

  The closer the center point is to the longitudinal center of the effective area, the more PPOS decreases. The PPOS increases as the center point moves to the border on both sides of the effective area.

  In addition or alternatively, the distance P2 between the center point and the fingertip is determined, and the upward inclination PF and the downward inclination PB are determined by the following equations (18) and (19).

  Any of the parameters described above are compared to critical values determined experimentally to assess image acceptability in the second determination step of step 230 (final determiner 340) of FIG. In the particular embodiment shown in FIG. 10, at step 232, the undulation R0 is compared to the undulation critical value Troll, at step 234, the upper slope PF is compared to the upper slope critical value TPF, and at step 236, the downward slope. PB is compared with the downward slope critical value TPB. If any one of the above critical values does not satisfy the condition, the image is discarded and another input image is received in step 202 (image input 302). Otherwise, the image selection step of step 242 sends the image to a well-known characteristic extraction analysis step.

  Accordingly, a system and method for determining acceptability of a fingerprint image for fingerprint recognition according to a preferred embodiment of the present invention addresses at least one of the above-mentioned problems, namely focus, quality and position. Can do. An acceptable fingerprint image is analyzed by a well-known method. For example, Patent Document 3 entitled “Mobile Handheld Code Reader and Print Scanner System and Method” or “System and Method for Fingerprint Image”. Patent Document 4 entitled “Enhancement Using Partitioned Least-Squared Filters” or Patent Document 5 entitled “Determining an Alignment Between Two (Fingerprint) Images” was used as a reference for this application.

  If for any reason the fingerprint image is discarded, a signal is output to the user indicating that the fingerprint image is unacceptable for fingerprint analysis. Such an indication is an indication that the fingerprint image has been discarded, or an indication of which critical value is associated with the decision that the fingerprint image has been discarded.

  Preferred embodiments of the present invention have been disclosed in this application. Even if a specific term is used, it should be construed as being used in a general sense and only for the purpose of describing the invention, and is intended for purposes of the present invention and claims. It should not be construed as being used to limit the scope of the invention described in the scope.

  For example, the process of the present invention is implemented in software, such as a product having a machine-accessible medium that includes data that causes a machine to perform a method for determining the acceptability of a fingerprint image. Accordingly, those skilled in the art will appreciate that various modifications and equivalent other embodiments are possible without departing from the scope of the invention as set forth in the claims.

  The present invention is applicable to technical fields related to mobile devices.

It is drawing which shows advancing of the light explaining formation of an optical image. 3 is a flowchart illustrating a process according to an embodiment of the present invention. It is a block diagram which shows embodiment of this invention. It is a graph which shows the blurring degree with respect to an object distance. 3 is a diagram illustrating three different images and their corresponding gradient distributions. 3 is a diagram illustrating three different images and their corresponding gradient distributions. 3 is a diagram illustrating three different images and their corresponding gradient distributions. 3 is a diagram illustrating three different images and their corresponding gradient distributions. 3 is a diagram illustrating three different images and their corresponding gradient distributions. 3 is a diagram illustrating three different images and their corresponding gradient distributions. It is a flowchart which shows embodiment of the 1st determination step in FIG. It is a flowchart which shows embodiment of the 1st determination step in FIG. 3 is a diagram illustrating a finger position estimation model according to an embodiment of the present invention. It is drawing which shows a finger outline extraction image. It is a flowchart which shows embodiment of the 2nd determination step in FIG.

Explanation of symbols

302 image input 304 divider 306 focus estimator 308 focus determiner 310 QI estimator 312 QI determiner 314 validity determiner 322 fingerprint pose analyzer 324 undulation estimator 326 tilt estimator 328 undulation determiner 330 tilt determiner 340 final Determinator 350 Fingerprint analyzer

Claims (19)

In a method for determining acceptability of a fingerprint image to be analyzed,
Dividing the image into a plurality of blocks;
Determining the focus for each block;
Determining the validity of a block that determines the block as a valid block if the focus is sufficiently well focused and has a repetitive pattern ; otherwise, the block as an invalid block;
Summing the number of valid blocks in the image;
Determining the effectiveness of the image to determine that the image is valid if the number of valid blocks is sufficient;
Determining whether the block includes a repetitive pattern includes determining a quality index that includes consistency and symmetry of the gradient distribution of the block .   The method of claim 1, wherein the number of effective blocks is sufficient if a ratio of the number of effective blocks to the number of multiple blocks exceeds a predetermined critical value. Determining the validity of the block comprises:
The method of claim 1 including the step of comparing the measured focus with a focus critical value determined from a blur model. The method according to claim 1 , wherein the quality index is a sum of a weight value of the consistency and the symmetry. If the image is valid,
Detecting a center point and an outline of the image;
Determining undulation and slope from the center point and the contour line;
2. The method of claim 1, further comprising selecting the image for subsequent processing when the undulation and the slope are sufficiently small. The step of determining the undulation degree includes:
Determining a left side distance between the center point and a left boundary of the contour line;
Determining a right distance between the center point and a right boundary of the contour line;
The method according to claim 5 , further comprising: determining an undulation that is proportional to an absolute value difference between the right distance and the left distance and a ratio of a sum of the right distance and the left distance. The step of determining the slope includes
Determining an upward slope;
6. The method of claim 5 , comprising determining a downward slope. The step of determining the slope includes
Determining a first distance from the center point to the upper boundary of the effective area;
Determining a second distance from the center point to the top of the contour line;
Determining a third distance from the center point to a lower boundary of the effective area; and
The upper gradient is proportional to the ratio of the first distance to said second distance, the declination degree according to claim 7, characterized in that in proportion to the ratio of the third distance to said second distance Method. The upper and lower boundaries of the effective area are:
9. The method according to claim 8 , wherein the width of the effective area is determined to be larger than half the width of the left boundary and the right boundary of the contour line in both the upper boundary and the lower boundary. The step of determining the slope includes
Determining a left side distance DL between the center point and a left boundary of the contour line;
Determining a right distance DR between the center point and a right boundary of the contour line;
Determining an upper distance DF from the center point to the upper boundary of the effective region;
Determining a lower distance DB from the center point to the lower boundary of the effective area;
Determining N to be greater than the ratio of the width and height of the effective area,
The method according to claim 5 , wherein the inclination is N− (DF + DB) / (DL + DR). The upper and lower boundaries of the effective area are:
The method according to claim 10 , wherein the width of the effective area is determined to be larger than half of the width of the left boundary and the right boundary of the contour line in both the upper boundary and the lower boundary.   The method of claim 1, further comprising generating the image using a camera. The method of claim 12 , wherein the camera is part of a portable device.   The method of claim 1, wherein determining the validity of the block comprises using a VMLOG (Variance-Modified-Lapalasian of Gaussian) method. The VMLOG method is:

15. The method of claim 14 , comprising at least one equation. A method for determining acceptability of a fingerprint image in a product having a machine accessible medium containing data that causes the machine to determine acceptability of a fingerprint image to be analyzed, comprising:
Dividing the image into a plurality of blocks;
For each block, determining whether the block is well focused;
Determining a state of the block that determines the block as a valid block if the block is well focused and has a repetitive pattern ; otherwise, the block is determined as an invalid block;
Summing the number of valid blocks in the image;
Determining that the image is valid if the number of valid blocks is acceptable;
Determining whether the block includes a repetitive pattern includes determining a quality index that includes consistency and symmetry of the gradient distribution of the block . The method of claim 16 , wherein the number of valid blocks is acceptable if the ratio of the number of valid blocks to the number of blocks exceeds a predetermined critical value. In an apparatus for determining acceptability of a fingerprint image to be analyzed,
A divider for dividing the image into a plurality of blocks;
For each block, a focus estimator that determines whether the block is well focused;
A focus determiner that compares the focus of the block with a focus critical value and determines the block as a valid block if the block is sufficiently focused; otherwise, the block is determined as an invalid block;
A quality index determinator that determines the state of a block that determines the block as a valid block if the block is well focused and has a repetitive pattern; otherwise, the block is determined as an invalid block;
A validity determiner that sums up the number of valid blocks in the image and determines that the image is valid if the number of valid blocks is acceptable. In an apparatus for determining the acceptability of a fingerprint image to be analyzed,
A detector for detecting a center point and an outline of the image;
A processor for determining a undulation and inclination from the center point and the contour line, and selecting the image for subsequent processing when the undulation and inclination are sufficiently small. The apparatus of claim 18 .

Images