HK1249645B - A method and device for acquiring an inclination value of an inclined image, a terminal and a storage medium - Google Patents

Description

Method, device, terminal and storage medium for obtaining tilt value of tilted image

Technical Field

The present invention relates to the field of image processing, and in particular to a method, device, terminal and storage medium for obtaining a tilt value of a tilted image.

Background Art

At present, when a rectangular image is tilted, if you want to obtain its tilt value, you can use the Hough line segment detection algorithm to detect and extract the boundary lines of the image. However, since the rectangular image includes boundary lines along the length and width directions of the image, and the boundaries of some rectangular images are not clear enough, multiple line segments will be extracted on the same boundary, and the background of some rectangular images is not pure enough, and there are impurity line segments. In this way, the Hough line segment detection algorithm will detect many boundary lines. Different boundary lines usually have different slope values. Even the slope values of multiple line segments extracted from the same boundary of a rectangular image will have some differences, so that the tilt value of the image cannot be uniquely determined.

Summary of the invention

In order to overcome the problem in the prior art that the tilt value of an image cannot be uniquely determined, embodiments of the present invention provide a method, device, terminal and storage medium for obtaining the tilt value of a tilted image.

In a first aspect, an embodiment of the present invention provides a method for obtaining a tilt value of a tilted image, wherein the tilted image is a rectangle, and the method for obtaining a tilt value includes:

Analyzing the oblique image to obtain coordinate information of a plurality of boundary lines of the oblique image;

According to each coordinate information, obtaining a first tilt value of each boundary line;

Get calibration value;

respectively calculating the difference between each of the first tilt values and the calibration value;

The first tilt value corresponding to the minimum difference is determined as the tilt value of the tilt image.

Preferably, the analyzing the oblique image to obtain coordinate information of a plurality of boundary lines of the oblique image includes:

Binarizing the oblique image to obtain a binary image;

The binary image is detected based on the Hough line segment detection algorithm to obtain coordinate information of the plurality of boundary lines.

Preferably, the tilted image is an ID card image; and obtaining the calibration value includes:

Parsing the ID card image and extracting all single-character regions, wherein the single-character region is an affine invariant region containing a single character;

Grouping all the single-character regions to obtain a plurality of single-character region groups; wherein the distance between any two adjacent single-character regions in the single-character region group is less than a first preset threshold;

Get the single-character region group with the longest length;

Obtain the inclination value of the connecting line segment of the first and last two single-character regions of the single-character region group with the longest length, and determine the inclination value of the connecting line segment as the calibration value.

Preferably, the step of grouping all the single-character regions to obtain a plurality of single-character region groups includes:

Get the first single character area;

Acquire all second-word single-character regions whose distances from the first single-character region are less than the first preset threshold;

until all n-th single-character regions whose distances from the n-1-th single-character region are less than the first preset threshold are obtained; wherein n is greater than 2, and the first single-character region, the second single-character region, and the n-th single-character region are different;

The first single character region, all the second single character regions, and up to all the nth single character regions are determined as a single character region group. Preferably, the oblique image includes a plurality of elements arranged in sequence, and the ordering direction of the plurality of elements is the same as the length direction or the width direction of the oblique image;

The obtaining of the calibration value comprises:

Analyze the oblique image to extract multiple element regions, where the element region is a region containing a single element;

Determine the adjacent area of two elements;

Obtain all second tilt values, where the second tilt value is a tilt value of a straight line connecting two adjacent element regions;

All the second tilt values are analyzed and calculated to obtain the calibration value.

Preferably, analyzing and calculating all the second tilt values to obtain the calibration value includes:

Clustering all the second tilt values to obtain multiple tilt value clusters;

Get the tilt value cluster with the largest weight;

The mean of the tilt value cluster with the largest weight is calculated to obtain the calibration value.

Preferably, the analyzing the oblique image and extracting multiple element regions comprises:

Analyze the oblique image and extract all the maximum stable extreme value areas;

Non-element regions are filtered out from all maximum stable extreme value regions to obtain a plurality of element regions.

In a second aspect, an embodiment of the present invention provides a device for obtaining a tilt value of a tilted image, wherein the tilted image is a rectangle, and the device for obtaining a tilt value comprises:

A parsing module, used for parsing the oblique image and obtaining coordinate information of a plurality of boundary lines of the oblique image;

A first tilt value acquisition module, used to analyze and calculate each coordinate information respectively to obtain a first tilt value of each boundary line;

A calibration value acquisition module, used for acquiring a calibration value;

A difference calculation module, used for respectively calculating the difference between each of the first tilt values and the calibration value to obtain a minimum difference;

A determination module is used to determine a first tilt value corresponding to the minimum difference as the tilt value of the tilt image.

In a third aspect, an embodiment of the present invention provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-mentioned methods for obtaining the tilt value of the tilted image when executing the computer program.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for obtaining the tilt value of the tilted image described in any one of the above items is implemented.

The embodiment of the present invention obtains the calibration value, compares the first tilt value of each extracted boundary line with the calibration value one by one, and finally determines the first tilt value corresponding to the minimum difference as the tilt value of the tilt image, so that the tilt value of the tilt image is uniquely determined.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

FIG1 is a schematic flow chart of a first embodiment of a method for obtaining a tilt value of a tilted image according to the present invention;

FIG2 is a schematic flow chart of a second embodiment of a method for obtaining a tilt value of a tilted image according to the present invention;

FIG3 is a schematic flow chart of an embodiment of a calibration value acquisition method of the present invention;

FIG4 is a flow chart of another embodiment of a calibration value acquisition method of the present invention;

FIG. 5 is a schematic structural diagram of an embodiment of a device for acquiring a tilt value of a tilted image according to the present invention.

DETAILED DESCRIPTION

In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

Please refer to FIG1 , which is a flow chart of a first embodiment of a method for obtaining a tilt value of a tilted image according to the present invention. The method comprises:

Step 101, parse the oblique image and obtain the coordinate information of multiple boundary lines of the oblique image. This step can be specifically performed by binarizing the oblique image to obtain a binary image, that is, a black and white image; the binary image is detected based on the Hough line segment detection algorithm to obtain the coordinate information of multiple boundary lines. The coordinate information can be the slope and intercept of the boundary line in the plane rectangular coordinate system (x, y), or the coordinates of the point corresponding to the boundary line in its parameter plane (k, b).

The binarization of the oblique image is to set the gray value of the pixel in the oblique image to the first value or the second value, that is, to present the entire oblique image with an obvious visual effect of only black and white. The binarization methods for the oblique image include the bimodal method, the iterative method, the P parameter method, etc. In addition to the listed binarization methods, there are many other binarization methods, which are not listed one by one in the disclosed embodiment. The detailed steps of the binarization method can refer to the relevant technology, and this embodiment is not specifically described. It should be noted that the first value and the second value can be set in advance, and the first value is greater than the second value. For example, the first value can be 255, 254, 253, etc., and the second value can be 0, 1, 2, etc. In order to more accurately obtain the contour of the straight line (i.e., the boundary line) to be detected in the oblique image, thereby improving the accuracy of the straight line detection, the first value can be 255, and the second value can be 0, which is not specifically limited in this embodiment.

Before detecting a binary image based on the Hough line segment detection algorithm, it is necessary to obtain the grayscale value of each pixel in the binary image, and based on the obtained grayscale value of each pixel, select a pixel whose grayscale value is a first value from the binary image area, and then determine the pixel coordinates of the selected pixel.

In a plane rectangular coordinate system (x, y), a straight line can be represented by the equation y=kx+b. For a certain point (x0, y0) on the straight line, y0=kx0+b, which represents a straight line in the parameter plane (k, b). Therefore, a pixel point in the binary image corresponds to a straight line in its parameter plane, and a straight line in the binary image corresponds to a point in the parameter plane. Therefore, this step can perform Hough transform on all pixel points with grayscale values of the first value on the binary image to obtain the straight line corresponding to each pixel point in the parameter plane; and determine the coordinates of the intersection points of multiple straight lines in the parameter plane, count the number of intersection lines corresponding to the intersection points, determine the intersection points whose number of intersection lines exceeds a preset number as the points corresponding to each boundary line in the parameter plane, and determine the coordinate information of the intersection points as the coordinate information of the boundary points. Specifically, the coordinates (k, b) of the intersection points are the slope and intercept of the boundary line.

Step 102, according to each coordinate information, obtain the first tilt value of each boundary line. The first tilt value can be a tilt angle or a slope. Specifically, when the first tilt value is a slope, the slope of each boundary line can be extracted from the coordinate information of each boundary line; when the first tilt value is a tilt angle, the slope k of each boundary line can be extracted from the coordinate information of each boundary line, and the slope is calculated to obtain the tilt angle arctank.

Step 103, obtaining a calibration value. The calibration value is used to filter or screen the first tilt values of each extracted boundary line, thereby obtaining a unique tilt value of the tilted image. There are many ways to obtain the calibration value.

In the first method, when the oblique image is a rectangle, the obtained multiple first oblique values can be clustered to obtain two clusters, and the variance of each cluster is less than the preset variance value; then the sum of the lengths of the line segments corresponding to the first oblique values in each cluster is calculated respectively; the maximum or minimum value of the two length sums obtained above is determined, and the mean of the cluster corresponding to the maximum or minimum value is used as the calibration value. Usually, two types of boundary lines are extracted: one is the boundary lines roughly parallel to the length direction of the oblique image, and the other is the boundary lines roughly parallel to the width direction; and the sum of the lengths of all boundary lines roughly parallel to the length direction is greater than the sum of the widths of all boundary lines roughly parallel to the width direction. Therefore, by using the mean of the cluster corresponding to the maximum or minimum value as the calibration value, the direction of the straight line corresponding to the calibration value can be determined, and the straight line direction is roughly parallel to the length direction or the width direction of the oblique image. If the mean of the cluster corresponding to the maximum value is used as the calibration value, then the tilt direction corresponding to the tilt value obtained in step 105 is the length direction of the tilted image; if the mean of the cluster corresponding to the minimum value is used as the calibration value, then the tilt direction corresponding to the tilt value obtained in step 105 is the width direction of the tilted image.

The second method is that if the oblique image includes multiple elements arranged in order, and the ordering direction of the multiple elements is the same as the length direction or the width direction of the rectangular oblique image, the oblique image can be parsed to extract multiple element regions, wherein the element region is a region containing a single element in the oblique image; two adjacent element regions are determined; all second inclination values are obtained, and the second inclination value is the inclination value of a straight line connecting two adjacent element regions; all second inclination values are analyzed and calculated to obtain calibration values.

The method of obtaining the calibration value using the second method will be described in detail below, so it will not be described here.

The third method is that if the oblique image is a photograph of an ID card, and the oblique image contains multiple characters, the oblique image can be parsed to extract multiple character regions, wherein the character region is an affine invariant region containing characters; determine all two adjacent character regions, wherein the distance between the two adjacent character regions is less than a preset distance value, and the preset distance value is less than or equal to the minimum line spacing of the characters; obtain all second inclination values, wherein the second inclination value is the inclination value of the straight line connecting the two adjacent character regions; average all the second inclination values to obtain the calibration value. Since the column spacing between the character strings corresponding to the address information and the number information on the ID card is less than the minimum line spacing of the characters on the ID card, the straight line connecting any two adjacent character regions is roughly parallel to the length direction of the oblique image, so that the inclination direction corresponding to the calibration value is the length direction of the oblique image. Among them, the method of parsing the oblique image and extracting multiple character regions is the same as the method of parsing the oblique image and extracting multiple element regions, which is specifically shown in the embodiment shown in Figure 3 and will not be repeated here.

The fourth method is, if the inclined image is a photograph of an ID card, then the ID card image is parsed to extract all single-character areas, where the single-character area is an affine invariant area containing a single character; all single-character areas are grouped to obtain multiple single-character area groups; wherein the distance between any two adjacent single-character areas in the single-character area group is less than a first preset threshold; the single-character area group with the largest length is obtained; the inclination value of the connecting line segment of the first and last two single-character areas of the single-character area group with the largest length is obtained, and the inclination value of the connecting line segment is determined as a calibration value.

Step 104 , respectively calculating the difference between each first tilt value and the calibration value to obtain a minimum difference.

Step 105: determine the first tilt value corresponding to the minimum difference as the tilt value of the tilt image.

The embodiment of the present invention obtains the calibration value, compares the first tilt value of each extracted boundary line with the calibration value one by one, and finally determines the first tilt value corresponding to the minimum difference as the tilt value of the tilt image, so that the tilt value of the tilt image is uniquely determined.

Please refer to FIG. 2, which is a flow chart of a second embodiment of a method for obtaining a tilt value of a tilted image of the present invention. In this embodiment, the tilted image comes from a photographed image of a tilted document. The method includes:

Step 201, obtaining a photographed image of a tilted document, wherein the photographed image includes a background image and a tilted image. The tilted document may be an identity card, a social security card, or a bank card.

Step 202: parse the captured image, remove the background image, and obtain an oblique image. Specifically, feature information of a preset oblique image may be obtained, and the feature information may be a shape feature or a color brightness feature; search for an image region matching the feature information in the captured image; and remove the region outside the image region (i.e., the background image) in the captured image, thereby obtaining an oblique image.

Step 203, determining whether the size of the tilted image is larger than the size of the tilted document.

Step 204: If the size of the tilted image is larger than the size of the tilted document, the size of the tilted image is set to the size of the tilted document. If the size of the tilted image is smaller than or equal to the size of the tilted document, the size of the tilted image is not set.

Step 205, parse the oblique image to obtain coordinate information of multiple boundary lines of the oblique image. This step can specifically be to perform binarization processing on the oblique image to obtain a binary image; and to detect the binary image based on the Hough line segment detection algorithm to obtain coordinate information of multiple boundary lines. The coordinate information can be the coordinate information of the two endpoints of the boundary line.

Step 206: Analyze and calculate each coordinate information to obtain a first tilt value of each boundary line. The first tilt value may be a tilt angle or a slope.

Step 207: Acquire the calibration value. The method for acquiring the calibration value has been described in detail above, so it will not be described in detail here.

Step 208: Calculate the difference between each first tilt value and the calibration value.

Step 209: determine the first tilt value corresponding to the minimum difference as the tilt value of the tilt image.

Since the tilted image is obtained based on the photographed image of the tilted document, the size of the tilted image will deviate from the size of the tilted document. In the process of boundary line extraction, the larger the size of the tilted image, the more difficult it is to extract the boundary line. Therefore, the embodiment of the present invention sets the size of the tilted image to the size of the tilted document when the size of the tilted image is larger than the size of the tilted document, thereby ensuring that the tilted image is not too large and will not affect the extraction of the boundary line.

Please refer to FIG. 3, which is a flow chart of an embodiment of the calibration value acquisition method of the present invention. This embodiment introduces the second method of obtaining the calibration value in detail. Among them, the oblique image includes multiple elements arranged in order, and the ordering direction of the multiple elements is the same as the length direction or width direction of the rectangular oblique image. The calibration value acquisition method specifically includes:

Step 301, analyzing the oblique image, extracting multiple element regions, wherein the element region is an affine invariant region containing elements. Step 301 includes:

301a, analyze the oblique image and extract all maximum stable extreme value regions. The maximum stable extreme value region is the most stable region obtained when the oblique image is binarized using different grayscale thresholds. Specifically, this step includes: obtaining a preset number of grayscale thresholds, using each grayscale threshold to binarize the oblique image, and obtaining a binarized image corresponding to each grayscale threshold; obtaining a region that maintains a stable shape in each binarized image corresponding to a preset grayscale threshold range, and obtaining a maximum stable extreme value region.

301b, determine the rectangular boundaries of each maximum stable extremum region. Specifically, this step includes: determining the outline of the maximum stable extremum region; according to the determined outline, obtaining the minimum circumscribed rectangle of the outline, thereby obtaining the rectangular boundary of the maximum stable extremum region. Among them, the minimum circumscribed rectangle refers to the maximum range of the two-dimensional shape represented by two-dimensional coordinates, that is, the rectangle whose lower boundary is determined by the maximum horizontal coordinate, the minimum horizontal coordinate, the maximum vertical coordinate, and the minimum vertical coordinate of each vertex of the given two-dimensional shape.

301c, filter out non-element regions from all maximum stable extreme value regions to obtain multiple element regions, wherein the non-element region is specifically the maximum stable extreme value region corresponding to the first rectangular boundary, and the element region is a region containing a single element. Since all the extracted maximum stable extreme value regions will include multiple non-element regions, they need to be filtered out to avoid interfering with the implementation of subsequent steps, which ultimately results in too large an error in the calibration value. Since each maximum stable extreme value region is an irregular region, it is not convenient to calculate its center point, nor is it convenient to remove the non-element region. Therefore, it is necessary to determine an external rectangular boundary for each maximum stable extreme value region to facilitate the calculation of the center point of the element region.

Specifically, 301c is as follows: detecting whether there is a first rectangular boundary. The first rectangular boundary is a rectangular boundary located inside other rectangular boundaries, a rectangular boundary with an area greater than a second preset threshold, or a rectangular boundary with an aspect ratio greater than a third preset threshold; if the first rectangular boundary is detected to exist, the maximum stable extreme value region corresponding to the first rectangular boundary is filtered out from all maximum stable extreme value regions. If the oblique image is an ID card image, the second preset threshold may be one quarter of the area of the oblique image, and if the element is a character, the third preset threshold may be 1.5.

Step 302, determine two adjacent element regions. Specifically, the distance between the two element regions can be obtained; determine whether the distance is less than the third preset threshold; if the distance is less than the third preset threshold, the two element regions are determined as two adjacent element regions. As a first embodiment, the distance between the two element regions can be the distance between the center points of the two element regions. As a second embodiment, the distance between the two element regions is the horizontal distance and/or the vertical distance of the two element regions in the preset coordinate system. When multiple elements in the image are arranged in multiple rows and columns, the third preset threshold is less than twice the element row spacing or twice the column spacing; or the third preset threshold is less than twice the element row spacing and less than twice the element column spacing. The two adjacent element regions determined include two element regions adjacent in the element sorting direction, and/or two element regions adjacent in the vertical direction. Among them, the vertical direction is a direction perpendicular to the element sorting direction. The number of straight line directions connecting two adjacent element regions (hereinafter referred to as adjacent directions) depends on the element row spacing, column spacing and the third preset threshold.

Step 303, obtain all second tilt values, where the second tilt value is the tilt value of the straight line connecting two adjacent element regions. The distance between two adjacent element regions is less than a preset distance value. The size of the preset distance value can be customized according to actual needs. In this embodiment, the preset distance value should be less than twice the row spacing or column spacing of the element.

Step 304, analyze and calculate all the second tilt values to obtain the calibration value. This step includes: 304a, clustering all the second tilt values to obtain multiple tilt value clusters; 304b, obtaining the tilt value cluster with the largest weight (i.e., the tilt value cluster with the largest number of tilt values); 304c, calculating the mean of the tilt value cluster with the largest weight to obtain the calibration value. Since the two adjacent element regions determined in step 302 include element regions adjacent along the length direction of the tilt image and element regions adjacent along the width direction of the tilt image, the calibration value can be uniquely determined by clustering and taking the mean of the tilt value cluster with the largest weight as the calibration value.

Among them, step 304a is specifically as follows: obtain a preset number; cluster all the second tilt values to obtain a preset number of tilt value clusters; calculate the variance of each tilt value cluster to obtain a minimum variance value; determine whether the minimum variance value is less than a first preset threshold; if the minimum variance value is greater than or equal to the first preset threshold, update the preset number, return to perform the step of clustering all the second tilt values to obtain a preset number of tilt value clusters, until the minimum variance value is less than the first preset threshold. The preset number is usually 2, and the method of changing the preset number is: add 1 to the current preset number value to obtain a new value, and assign the new value to the preset number. If the tilted image is an image of an ID card, then the first preset value is 200.

Specifically, the step of clustering all the tilt values to obtain a preset number of tilt value clusters includes: arbitrarily selecting a preset number of tilt values from all the tilt values as initial cluster centers; and for the remaining tilt values, according to their similarity with these initial cluster centers (i.e., the distance from the initial cluster centers), respectively assigning them to the clusters represented by the initial cluster centers that are most similar to them, to obtain a preset number of clusters.

Please refer to Figure 4, which is a flow chart of another embodiment of the calibration value acquisition method of the present invention. This embodiment introduces the fourth calibration value acquisition method in detail.

Step 401, parse the ID card image and extract multiple single-character regions. The ID card image includes multiple characters and multiple lines of characters arranged in order, and the characters can be text or numbers. Among them, the single-character region is an affine invariant region containing a single character. Specifically, based on the regional feature extraction algorithm, all affine invariant regions in the ID card image can be extracted, and non-single-character regions (i.e., affine invariant regions that do not have all the features of a single character) can be eliminated therefrom to obtain a single-character region.

Step 402, obtaining the first single-character region. For the first obtained single-character region group, the first single-character region can be randomly selected from all extracted single-character regions; for the non-first obtained single-character region group, the first single-character region needs to be randomly selected from the remaining single-character regions. The remaining single-character regions refer to the multiple single-character regions obtained after removing the already obtained single-character region group from all single-character regions.

Step 403, obtain all second single-character regions whose distances to the first single-character region are less than a first preset threshold. Among them, the distance between any two adjacent single-character regions in the single-character region group is less than the first preset threshold. Furthermore, in order to divide all single-character regions located in the same row on the ID card image into the same group as much as possible, the distance between any single-character region in any single-character region group and any single-character region in another single-character region group is greater than or equal to the first preset threshold. The first preset threshold has been introduced in detail above, so it will not be repeated here.

For single-character regions in the same row, the single-character region in the middle usually has at least two single-character regions whose distances are less than the first preset threshold. Therefore, in order to obtain all single-character regions in the same row, it is necessary to obtain all second single-character regions.

Step 404, until all the n-th single-character regions whose distance to the n-1th single-character region is less than the first preset threshold are obtained. This step can be understood as taking the newly obtained single-character region as the benchmark in turn, obtaining all the single-character regions whose distance to the benchmark is less than the first preset threshold, until the n-th single-character region. Wherein, n is greater than or equal to 2. In order to ensure that the single-character region is not repeatedly obtained, the first single-character region, the second single-character region, and the n-th single-character region are different from each other. Wherein, n can be a preset value, for example, n can be 10. In addition, n can also be an indefinite value. In order to divide all the single-character regions located in the same row on the ID card image into the same group as much as possible, step 404 includes: obtaining all the n-th single-character regions whose distance to the n-1th single-character region is less than the first preset threshold, until the acquisition of the n+1th single-character region fails. Wherein, the n+1th single-character region is different from the first single-character region, the second single-character region, and the n-th single-character region.

Specifically, step 404 includes: obtaining all third single-character regions whose distance from the second single-character region is less than the first preset threshold; obtaining all fourth single-character regions whose distance from the third single-character region is less than the first preset threshold... until obtaining all n-th single-character regions whose distance from the n-1-th single-character region is less than the first preset threshold. In the process of implementing this step, since the distances between the third single-character region and the first character region and the second character region are less than the first preset threshold, the third single-character region obtained will be repeated with the first single-character region obtained previously. Therefore, in the process of implementing step 304, in order to ensure that the currently obtained single-character region is different from all the previously obtained single-character regions, specifically,

Specifically, after the step of acquiring all the m-th single-character regions whose distance from the m-1th single-character region is less than the first preset threshold, it can be determined whether the newly acquired m-th single-character region overlaps with all the previously acquired single-character regions (including all the first single-character regions, the second single-character regions up to the m-th single-character region). If there is overlap, the process is terminated. If there is no overlap, the next step is executed: taking the newly acquired single-character region as a reference (i.e., the m-th single-character region), all the m+1-th single-character regions whose distance from the reference (m single-character region) is less than the first preset threshold are acquired. Wherein, m is less than or equal to n-1.

Step 405: determine the first single-character region, the second single-character region, up to the nth single-character region as a single-character region group.

Step 406, obtaining the single-character region group with the largest length. This step specifically includes: calculating the length of each single-character region group respectively, and selecting the single-character region group with the largest length.

The length calculation of the single-character area group includes: obtaining the length of the connecting line segment of the first and last single-character areas of the single-character area group; and determining the length of the connecting line segment as the length of the single-character area group. Among them, the step of obtaining the length of the connecting line segment of the first and last single-character areas of the single-character area group includes: obtaining the distance between any two single-character areas in the current single-character area group, and filtering out the maximum distance; and determining the maximum distance as the length of the connecting line segment of the first and last single-character areas of the current single-character area group. Furthermore, the step of filtering out the single-character area group with the largest length includes: filtering out the maximum length from the lengths of all the connecting line segments obtained, and determining the single-character area group corresponding to the maximum length as the single-character area group with the largest length.

Step 407, obtain the inclination value of the connecting line segment of the first and last single-character areas of the single-character area group with the longest length, and determine the inclination value of the connecting line segment as the calibration value. The connecting line segment is a line segment with the reference points of the first and last single-character areas as endpoints. The reference point is specifically the geometric center point of the single-character area. By obtaining the coordinates of the two endpoints of the connecting line segment, the inclination value of the connecting line segment can be obtained. The inclination value can be an inclination angle or a slope. Since the shapes of the characters are different, this will result in that the geometric center points of the extracted single-character areas cannot be located on the same straight line at the same time. Therefore, by obtaining the inclination value of the connecting line segment of the first and last single-character areas of the single-character area group with the longest length, the error can be minimized to be close to the inclination value of the real ID card image.

It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

The above is a detailed introduction to the method for obtaining the tilt value of a tilted image of the present invention. The following is a further explanation of the device for the method.

Please refer to FIG. 5 , which is a schematic diagram of the structure of an embodiment of a device for obtaining a tilt value of a tilted image according to the present invention.

The tilt value acquisition device comprises: an analysis module 501 , a first tilt value acquisition module 502 , a calibration value acquisition module 503 , a difference calculation module 504 and a determination module 505 .

The analysis module 501 is used to analyze the oblique image and obtain the coordinate information of multiple boundary lines of the oblique image. This step can specifically be to perform binarization processing on the oblique image to obtain a binary image; and to detect the binary image based on the Hough line segment detection algorithm to obtain the coordinate information of multiple boundary lines.

The first tilt value acquisition module 502 is used to analyze and calculate each coordinate information to obtain a first tilt value of each boundary line. The first tilt value can be a tilt angle or a slope.

The calibration value acquisition module 503 is used to acquire the calibration value. The method of acquiring the calibration value has been described in detail above, so it will not be repeated here. If the oblique image includes a plurality of elements arranged in sequence, and the ordering direction of the plurality of elements is the same as the length direction or the width direction of the rectangular oblique image, then the calibration value acquisition module 503 includes an element extraction unit, an adjacent element determination unit, a second oblique value acquisition unit and a calibration value calculation unit.

The element extraction unit is used to analyze the tilted image and extract multiple element regions, wherein the element region is an affine invariant region containing elements. The adjacent element determination unit is used to determine two adjacent element regions. The second tilt value acquisition unit is used to acquire all second tilt values, where the second tilt value is the tilt value of a straight line connecting two adjacent element regions. The calibration value calculation unit is used to analyze and calculate all second tilt values to obtain calibration values.

The difference calculation module 504 is used to calculate the difference between each first tilt value and the calibration value.

The determination module 505 is configured to determine the first tilt value corresponding to the minimum difference as the tilt value of the tilt image.

The embodiment of the present invention obtains the calibration value, compares the first tilt value of each extracted boundary line with the calibration value one by one, and finally determines the first tilt value corresponding to the minimum difference as the tilt value of the tilt image, so that the tilt value of the tilt image is uniquely determined.

In addition, the device for obtaining the tilt value of the tilted image also includes a captured image acquisition module, a background removal module, a judgment module and a setting module. The captured image acquisition module is used to obtain a captured image of the tilted document, wherein the captured image includes a background image and a tilted image. The tilted document may specifically be an identity card, a social security card or a bank card, etc. The background removal module is used to parse the captured image, remove the background image, and obtain a tilted image. The judgment module is used to determine whether the size of the tilted image is larger than the size of the tilted document. The setting module is used to set the size of the tilted image to the size of the tilted document if the size of the tilted image is larger than the size of the tilted document. Specifically, the parsing module 501 is used to parse the tilted image set by the setting module to obtain the coordinate information of multiple boundary lines of the tilted image.

Since the tilted image is obtained based on the photographed image of the tilted document, the size of the tilted image will deviate from the size of the tilted document. In the process of boundary line extraction, the larger the size of the tilted image, the more difficult it is to extract the boundary line. Therefore, the embodiment of the present invention sets the size of the tilted image to the size of the tilted document when the size of the tilted image is larger than the size of the tilted document, thereby ensuring that the tilted image is not too large and will not affect the extraction of the boundary line.

An embodiment of the present invention further provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for obtaining the tilt value of the tilted image when executing the computer program.

An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for obtaining the tilt value of the tilted image is implemented.

A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

As described above, one or more implementation methods are provided in combination with specific contents, and the specific implementation of the present invention is not limited to these descriptions. Any similarity or similarity with the method, structure, etc. of the present invention, or any technical deduction or replacement based on the concept of the present invention, shall be regarded as the protection scope of the present invention.

Claims (6)

1. A method for obtaining a tilt value of a tilted image, wherein the tilted image is a rectangle, and the method for obtaining a tilt value comprises: Analyzing the oblique image to obtain coordinate information of a plurality of boundary lines of the oblique image; According to each of the coordinate information, obtaining a first tilt value of each of the boundary lines; Get calibration value; respectively calculating the difference between each of the first tilt values and the calibration value to obtain a minimum difference; Determine a first tilt value corresponding to the minimum difference as the tilt value of the tilt image; The oblique image includes a plurality of elements arranged in sequence, and the ordering direction of the plurality of elements is the same as the length direction or the width direction of the oblique image; The obtaining of the calibration value comprises: Analyze the oblique image to extract multiple element regions, where the element region is a region containing a single element; Determine the adjacent area of two elements; Obtain all second tilt values, where the second tilt value is a tilt value of a straight line connecting two adjacent element regions; Analyzing and calculating all the second tilt values to obtain the calibration value; The analyzing and calculating all the second tilt values to obtain the calibration value includes: Clustering all the second tilt values to obtain multiple tilt value clusters; Get the tilt value cluster with the largest weight; Calculating the mean of the tilt value cluster with the largest weight to obtain the calibration value; Wherein, all the second tilt values are clustered to obtain a plurality of tilt value clusters, including: Obtain a preset number; cluster all second inclination values to obtain a preset number of inclination value clusters; calculate the variance of each inclination value cluster to obtain a minimum variance value; determine whether the minimum variance value is less than a first preset threshold; if the minimum variance value is greater than or equal to the first preset threshold, update the preset number, and return to execute the step of clustering all second inclination values to obtain a preset number of inclination value clusters, until the minimum variance value is less than the first preset threshold. 2. The method for obtaining the tilt value of an oblique image according to claim 1, wherein the step of analyzing the oblique image to obtain coordinate information of a plurality of boundary lines of the oblique image comprises: Binarizing the oblique image to obtain a binary image; The binary image is detected based on the Hough line segment detection algorithm to obtain coordinate information of the plurality of boundary lines. 3. The method for obtaining the tilt value of an oblique image according to claim 1, wherein the step of analyzing the oblique image and extracting a plurality of element regions comprises: Analyze the oblique image and extract all the maximum stable extreme value areas; Non-element regions are filtered out from all maximum stable extreme value regions to obtain a plurality of element regions. 4. A device for obtaining a tilt value of a tilted image, wherein the tilted image is rectangular, and the device for obtaining a tilt value comprises: A parsing module, used for parsing the oblique image and obtaining coordinate information of a plurality of boundary lines of the oblique image; A first tilt value acquisition module, used to analyze and calculate each coordinate information respectively to obtain a first tilt value of each boundary line; A calibration value acquisition module, used for acquiring a calibration value; A difference calculation module, used for respectively calculating the difference between each of the first tilt values and the calibration value to obtain a minimum difference; A determination module, configured to determine a first tilt value corresponding to the minimum difference as a tilt value of the tilt image; The calibration value acquisition module is specifically used for: Analyze the oblique image to extract multiple element regions, where the element region is a region containing a single element; Determine the adjacent area of two elements; Obtain all second tilt values, where the second tilt value is a tilt value of a straight line connecting two adjacent element regions; Analyzing and calculating all the second tilt values to obtain the calibration value; The analyzing and calculating all the second tilt values to obtain the calibration value includes: Clustering all the second tilt values to obtain multiple tilt value clusters; Get the tilt value cluster with the largest weight; Calculating the mean of the tilt value cluster with the largest weight to obtain the calibration value; Wherein, all the second tilt values are clustered to obtain a plurality of tilt value clusters, including: Obtain a preset number; cluster all second inclination values to obtain a preset number of inclination value clusters; calculate the variance of each inclination value cluster to obtain a minimum variance value; determine whether the minimum variance value is less than a first preset threshold; if the minimum variance value is greater than or equal to the first preset threshold, update the preset number, and return to execute the step of clustering all second inclination values to obtain a preset number of inclination value clusters, until the minimum variance value is less than the first preset threshold. 5. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for obtaining the tilt value of a tilted image as described in any one of claims 1 to 3 when executing the computer program. 6. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the method for obtaining the tilt value of a tilted image according to any one of claims 1 to 3 is implemented.