KR20170095062A - A Method Of Providing For Searching Footprint And The System Practiced The Method - Google Patents
A Method Of Providing For Searching Footprint And The System Practiced The Method Download PDFInfo
- Publication number
- KR20170095062A KR20170095062A KR1020160016573A KR20160016573A KR20170095062A KR 20170095062 A KR20170095062 A KR 20170095062A KR 1020160016573 A KR1020160016573 A KR 1020160016573A KR 20160016573 A KR20160016573 A KR 20160016573A KR 20170095062 A KR20170095062 A KR 20170095062A
- Authority
- KR
- South Korea
- Prior art keywords
- image
- contrast
- pattern
- comparative
- similarity
- Prior art date
Links
Images
Classifications
-
- G06F17/30271—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G06F17/30256—
-
- G06F17/3028—
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a method for providing a footprint search data, which provides objective data on the similarity of a comparison footprint and a comparative footprint stored in advance in searching for footprints similar to a comparative footprint, The present invention relates to a method for providing stitch search data.
Description
The present invention relates to a method for providing a footprint search data, and more particularly, to a method for providing footprint search data by providing objective data for searching for footprints so that similar footprints can be retrieved quickly and more accurately.
Recently, as the intelligence of crime has become more advanced, such as not leaving fingerprints, it is difficult to erase it easily, and there is an increasing demand for analysis of footprints that are likely to leave traces. Footprints are traces of shoe marks left on the floor or floor of a building by a criminal.
It is helpful to identify the suspects and suspects by promptly searching for the remaining footsteps in the crime scenes as well as in the strong cases and providing information about the types of shoes the criminal wears. Similar to DNA search, it is also helpful to identify the same crime and crime through the stigma.
FIG. 1 shows an example of a footstep, FIG. 2 shows a footstep DB, FIG. 3 schematically shows a contrast footstep and a contrast footstep, FIG. FIG. 5 schematically shows an example of a pattern DB, and FIG. 6 shows a system for searching a footstep for a conventional footstep search.
In the following description, "comparative streak" means a footstep left on the scene of the incident, and "contrast footstep " means a footstep stored in the storage unit to obtain information about a comparative footstep. The contrast footsteps are obtained from the shoes distributed on the market. For example, an ink or the like is applied to the bottom of a shoe which is distributed on the market, and is stamped on a paper or the like, shot, and obtained as a contrasting image, and stored in a storage unit.
As shown in FIG. 6, an apparatus for searching for a footprint similar to a comparative footprint includes a control unit, a storage unit connected to the control unit and storing a footprint database of a plurality of contrast shakes, An input unit (e.g., a keyboard and a mouse) as input means, a display unit connected to the control unit and displaying a contrast group or a comparison group image, and an output unit connected to the control unit and outputting a search result. The storage unit also stores a program that is executed to compare the comparison unit and the contrast unit. The storage unit may be a server communicating with the control unit.
Figure 1 shows an example of a comparative footprint image. The contrast streak image is also shown in Fig. The family trace image is saved as a picture file (eg bmp file) by taking a picture or the like.
The contrasted root image is binarized, and the binarized image may be displayed on the display unit. Since the contents of the binarization are conventionally known, the description of the binarization process is omitted.
As shown in FIG. 1, the
As shown in Fig. 2, the footprint database for the contrast footsteps comprises a shoe DB and a pattern DB. The shoe DB may include the name of the shoe model, the image of the shoe under the shoe (bottom) image, the pattern number, and the like. The pattern DB includes the pattern number and the pattern name. The footprint of each contrast footprint is stored in the storage.
As shown in FIG. 3, the images of the comparative footprints obtained by a method such as shooting on the spot are stored in the storage unit, and information such as the shoe model name is obtained by comparing with a plurality of contrast footprints.
When the user executes a program to search for a contrast streak image similar to the comparative striae image through the input unit, the pattern number and the pattern shape are displayed on the display unit. The user selects the pattern through the input unit, selects the pattern position, . A comparative strain image is displayed on one side of the display portion, and a pop-up window for inputting a glyph as shown in Fig. 5 is displayed. The area for the glyph input may be displayed with the comparative footprint image on one side of the comparative footprint image.
When the user selects the shape and position of the pattern through the input unit, the selected information is stored in the storage unit, and compared with the pattern DB (pattern shape and pattern position) of the contrast wave stored in the storage unit, The footprint image is displayed on the display side-by-side with the comparative footprint image. When there are plural contrast streaks, images of a plurality of contrast streaks are sequentially displayed in parallel with the comparison streak image according to a user's input command.
The method of searching existing footsteps is to judge whether the position of the footwear is in the middle, middle, or the bottom of the circle, rectangle, stripes, horseshoe which are predefined patterns in the footprints of analysts (investigators) And input the pattern and position in the shoe pattern database marketed in Korea.
However, the positions of the top, middle, and bottom are very fragmented, and accuracy is high only when the actual input is sufficient. If there is not enough input data, there are many kinds of shoes to be searched, and there is a problem that analysts have to perform secondary search with eyes. In addition, the footprint of a crime scene is rather blurry than a clean photograph, and it is difficult to find three or more patterns. Therefore, conventional search techniques are often ineffective and perform secondary search with a lot of human data.
SUMMARY OF THE INVENTION It is an object of the present invention to provide a method for providing a footprint search data to search for similar footprints more quickly and accurately by providing comparative footprints against similarities of comparative footprints.
In accordance with another aspect of the present invention, there is provided an image processing apparatus including a control unit, a storage unit connected to the control unit and storing a footprint database of a plurality of contrast shafts, a display unit connected to the control unit, Executed by a system consisting of an input section serving as a command input means; Wherein the footprint DB comprises a footprint DB for at least one pattern included in the shoe DB and the contrast footprint image, the footprint DB includes a contrast footprint image and a shoe model name, and the flint DB includes each pattern included in the contrast footprint image A contour vector for a pixel outside the glyph of the contour vector, an autocorrelation function value (ACF) computed from contour vectors, and a norm function value;
Calculating a contour vector for outline pixels of at least one of the glyphs included in the comparison group image stored in the storage unit; calculating an autocorrelation function value for each glyph from the contour vector; Calculating a norm function value for each of the patterns by comparing an autocorrelation function value of each contrast stance with an autocorrelation function value and a norm function value of the comparison group and a normal scalar value R of a norm function value, And displaying a normal scalar value of each contrast zone on the display; The outline is a closed loop, and the contour vector of each pattern provides a method of providing a sketch search data, which is an incremental complex vector of outline pixels forming a pattern.
In the above, the autocorrelation function value (ACF) for each pattern of the contrast group or the comparison group is calculated as a shift scalar sum of the incremental complex vector with respect to the outer pixel constituting each pattern
The function value (AC ACF)) for each pattern of the contrast streak or the comparison streak is the scalar sum of the incremental complex vector for the outer vector forming each glyph
The normal color value (R)
Where ACF (n) is a shift scalar sum computed for the glyphs contained in the comparison root image, ACF '(n) is the contour vector for the contour pixel, (N) is the computed function value for the pattern included in the comparative grouping image, and | ACF '(n) | is the contrast value of the contrasting sketch image Is a genomic function value computed for a glyph included in the glyph.
In the above, the autocorrelation function values of the contrast masks and the normal scalar set values of the normal scalar values (R) of the norm function values are stored in the storage unit for the autocorrelation function value and the norm function value of the comparative group, The shoe DB for the contrast shoe is displayed on the display unit together with the normal scalar value R only when the normal scalar value R is larger than the normal scalar setting value stored in the storage unit.
3. The method according to
In the above, the class correlation function value (ICF) for each pattern of contrast streaks and each pattern of comparative group shakes
In the above equations, v is a contour vector for the outline pixels of each pattern included in the comparative group image, v 'is a contour vector for the contour pixels of each pattern included in the contrast pattern image, and k is the number of contour pixels to be.
Before calculating the autocorrelation function value, the norm function value, and the class correlation function value, the number of outline pixels for each pattern included in the comparison pattern image and the contrast pattern image included in the comparison group image is counted and stored in the storage unit ; The autocorrelation function value, the norm function value, and the class correlation function value are calculated for each pattern included in the contrast pattern images having the same number of pixels and the number of pixels included in the comparison group image.
In the above, the shoe DB includes data of the relative footstep origin position of the contrast streak image calculated in the contrast streak image, contrast center data of the center of each pattern included in the contrast streak images, Further comprising contour pedestal gauging distance data that is a distance between the center point of the contrast stair pattern;
The method for providing the stigma retrieval data further includes a step of computing the relative distance of each pattern included in the comparative group image, and a step of providing distance similarity data;
Wherein the step of computing the comparative core stiffness includes calculating the origin position (comparison root image origin position data) of the comparative rootstock image in the comparative rootstock image, calculating the central position of each pattern included in the comparative rootstock image Calculating a distance between an origin position of the comparison group image and each center position of the comparison group (comparative arch span distance data);
The distance similarity data providing step may include a step of calculating a distance similarity degree of each pattern included in the contrast pattern images for each pattern included in the comparison group image and a step of calculating maximum distance similarity values for each pattern included in the comparison group image And calculating an average distance similarity value for the derived maximum distance similarity values;
Wherein the display unit displays the average distance similarity of each of the contrast shakes relative to the comparison shoe,
The distance similarity (tn, tn ') for each pattern included in the comparative footprint images for each pattern included in the comparative footprint image is
Where dn is the pattern distance of the angle included in the comparative footprint image and dn 'is the pattern distance of each pattern included in the contrast footprint images.
In the above description, the shoe DB includes the origin position data of the contrast shoe images calculated from the contrast shoe images, the contrast center shoe center data which is the center position of each shoe included in the contrast shoe image, Further comprising angle data;
Wherein the method for providing the footstep search data further includes a step of calculating a relative footstep angle and a step of providing the degree of similarity data, the flint angle being calculated for each pattern included in the comparative footstep image;
The step of computing the comparative core stiffness angle includes a step of calculating an origin position (comparison root image origin position data) of the comparative grouping image in the comparative grouping image, a step of calculating a center position of each pattern included in the comparative grouping image Calculating a center angle of each of the patterns relative to an origin position of the comparative group image (comparison pattern image angle data);
The step of providing the degree of similarity data may include the steps of calculating the degree of similarity of each pattern included in the contrasted grouped images for each pattern included in the comparative grouped image, and calculating the maximum degree of similarity values for each pattern included in the comparative grouped image And calculating an average degree of similarity value for the derived maximum degree of similarity values;
Wherein the display unit displays the average degree of similarity of the respective contrast shots with respect to the comparative group,
The angle similarity (? N,? N ') of each pattern included in the contrast pattern images for each pattern included in the comparative group image
Where rn is the pattern angle of each pattern included in the comparative footprint image, and rn 'is the pattern angle of each pattern included in the contrast footprint image.
In the above, the shoe DB stores data of the relative footstep origin position of the contrast streak image calculated from the contrast streak images, the contrast center staple center data of the center of each pattern included in the contrast streak image, Further comprising contour pedestal crossing distance data which is a distance between center point positions of the patina stones and contrast pedestal crossing angle data which are angles of center strokes of the contrast stones;
The method for providing the stigma retrieval data includes a step of computing the relative angle of each of the patterns included in the comparative group image, a step of computing the relative distance of each pattern included in the comparative group image, And a similarity data providing step;
The step of computing the comparative core stiffness angle includes a step of calculating an origin position (comparison root image origin position data) of the comparative grouping image in the comparative grouping image, a step of calculating a center position of each pattern included in the comparative grouping image Calculating a center angle of each of the patterns relative to an origin position of the comparative group image (comparison pattern image angle data);
Wherein the step of computing the comparative core stiffness comprises the step of calculating a distance between the origin position of the comparative grouping image and the center position of each pattern (comparative legend image pattern distance data);
Wherein the similarity data providing step includes the steps of calculating distance similarities of the patterns included in the contrasting stone images for the respective patterns included in the comparison group image,
Calculating the degree of similarity of each of the patterns contained in the contrasted root image for each pattern included in the comparative group image,
Calculating a degree of similarity of each pattern included in the contrasted root image for each pattern included in the comparison root image from the distance similarity and the degree of similarity;
Calculating maximum similarity values for each pattern included in the comparison group image, and calculating an average similarity value for the derived maximum similarity values;
Wherein the display unit is configured to display an average similarity of each of the contrast shakes with respect to the comparative group,
The angle similarity (? N,? N ') for each pattern included in the contrast pattern images for each pattern included in the comparative group image
Where rn is the gyration angle of the angle included in the comparative group image, and rn 'is the gyration angle of each glyph included in the contrast group image;
The distance similarity (tn, tn ') for each pattern included in the contrast image for each pattern included in the comparative group image is
Where dn is the gyration distance of the angle included in the comparative group image, dn 'is the gyration distance of each glyph included in the contrast group image,
The similarity (Sn, Sn ') of each pattern included in the contrast image for each pattern included in the comparative group image
According to the method of providing the footstep search data according to the present invention, objective numerical data for judging similarity can be provided in searching for a contrast streak similar to the comparative footstep, so that the user can more precisely and quickly compare the contrast streak It is possible to search and obtain information about the comparative streak.
Figure 1 shows an example of a footprint image,
2 is a view for explaining a footprint DB stored in a storage unit,
FIG. 3 is a diagrammatic representation for the comparison of the comparative and contrast zones,
FIG. 4 is a schematic view of a conventional footstep retrieval process,
Fig. 5 schematically shows an example of a pattern DB,
FIG. 6 shows a footprint search system for searching for a footprint,
FIG. 7 is a flowchart of a method for providing footprint search data according to the present invention,
8 is an enlarged view of the pixels of the pattern included in the footprint image,
FIG. 9 is a drawing showing the glyph pixels in numerical form,
FIG. 10 shows pixels of a pattern derived from internal pixels and outline pixels among the pixels forming the pattern,
FIG. 11 shows a pixel of a pattern from which outline pixels are derived out of the pixels forming the pattern,
12 and 13 are diagrams for explaining the outline vector derivation for the outline pixel among the pixels forming the pattern.
Hereinafter, a method for providing the present invention's footstep search data will be described in detail with reference to the drawings. In the description, the contents of the conventional known art are not described.
FIG. 7 is a flowchart of a method for providing a footprint search data according to the present invention, FIG. 8 is an enlarged view of pixels of a pattern included in a footprint image, FIG. 9 is a numerical representation of a pattern pixel, Fig. 10 shows the pixels of the pattern from which the inner pixels and the outline pixels are derived, and Fig. 11 shows the pixels of the pattern from which the outer pixels are derived from among the pixels forming the pattern. Figs. 12 and 13 Is used to illustrate the derivation of the contour vector for the outer pixel among the pixels forming the pattern.
The program in which the method for providing the footstep search data according to the present invention is executed is stored in the storage unit of the computer apparatus as shown in Fig. 6 and is executed by the execution instruction of the user. The user selects the comparative image of the staple stored in the storage unit and executes the program to execute the method of providing the staple image data.
As shown in FIG. 7, the method for providing the pedestrian detection data of the present invention includes a step (ST-110) of selecting a pedestrian trace image (comparative pedestrian image) and a step A step (ST-120, deriving a feature vector), a step (ST-130, a complex vector derivation) of deriving a contour vector from outline pixels for each glyph, a function (ST-140, operation step), and a step (ST-150) in which the calculated function values are displayed on the display unit. The calculated function values are displayed on the display unit, so that the user can obtain more objective numerical data, and can quickly and accurately search for the contrast streak image similar to the comparative streak image.
The comparative footprint image and the contrast footprint image are binarized and stored in the storage unit, and displayed on the display unit by a user's input command. The comparative footprint image and the contrast footprint image displayed on the display unit include a plurality of patterns as shown in FIG.
Describes the calculation process for the comparison pattern image and the pattern included in the contrast pattern image. Hereinafter, a process of calculating a pattern included in the comparative group image will be described, and a description of the contrast group will be omitted for the same part in the process of calculating the pattern included in the contrast group image.
FIG. 8 shows an example of a pattern included in the comparative grouping image, wherein the comparative grouping image is binarized and stored, and the pixels forming the comparative grouping image are R = 0, G G, and B = 255, G = 255, and B = 255 in the pixel color information (R, G, B) Quot; background pixel " hereinafter). FIG. 9 shows the background pixel as 0 and the glyph pixel as 1.
The outline pixels constituting the pattern are derived by the user clicking the outline pixels among the pattern pixels in each pattern of the comparison group image through the input unit and storing the outline pixel information (abscissa, ordinate) for each pattern in the storage unit It is possible.
The process of deriving the outline pixels constituting the pattern by the execution of the program will be described as an example.
First, when the program in which the method of providing the footprint search data is executed is executed so that all the four directions of the glyph pixels are the glyph pixels, the pixel color information (R, G, B) G = 255, and B = 255 to be displayed. The pattern consisting of the pixels as shown in FIG. 8 is displayed by changing a part of the glyph pixels to pixels having the same color information as the background pixels, as shown in FIG.
The maximum horizontal coordinate value, the maximum vertical coordinate value, and the minimum vertical coordinate value are derived from the coordinate values of the glyph pixels of the pixels as shown in FIG. 10, and the pixel Pm1 (Hereinafter referred to as " minimum pixel ") is derived and stored in the storage unit. When there are a plurality of pixels having a minimum sum of coordinate values, a pixel having the minimum vertical coordinate value (y coordinate value) becomes a minimum pixel and is stored in the storage unit together with the coordinate value information.
The distance between the glyph pixel and the minimum pixel Pm1 shown in Fig. 10 is calculated, and the pixel having the closest distance to the minimum pixel Pm1 is derived and stored in the storage unit. When there are two pixels having the same distance from the minimum pixel Pm1, a pixel having a large horizontal coordinate value (x coordinate value) is derived as a pixel (Ps1: first pixel) closest to the minimum pixel Pm1 and stored .
In the same way, a pixel closest to the first pixel Ps1 is derived and is executed until the x coordinate value of the pixel derived in this way becomes the maximum abscissa value. When the x coordinate value of the derived pixel becomes the maximum abscissa value, a pixel having the longest ordinate value (y coordinate value) among the closest distance pixels is derived and stored in the storage unit. (Y coordinate value) of the derived pixel is derived to be the maximum ordinate value and stored in the storage unit in order. When the derived ordinate value of the pixel is the maximum ordinate value, a pixel having the smallest abscissa value among the pixels having the closest distance among neighboring pixels is derived, and the information is stored in the storage unit. The pixel is derived and stored in the storage unit. When the derived coordinate value becomes the minimum abscissa value, a pixel having the smallest vertical coordinate value among the pixels having the closest distance is derived, and the information is stored in the storage unit, and until the pixel having the vertical ordinate value is derived And is stored in the storage unit. As shown in FIG. 10, when the color information (R, G, B) of the glyph pixels except for the pixels stored in the storage unit among the pixels constituting a certain pattern is changed to R = 255, G = 255, As shown in FIG. Outline pixels for a certain pattern are derived.
In this manner, the outline pixel information of one or more glyphs included in the comparative group image is stored in the storage unit. Likewise, outline pixel information for one or more glyphs included in the contrast group images is stored in the storage unit. Outer pixels form a closed circuit.
FIG. 12 exemplifies the outer pixels P1 to P10 constituting the pattern. In Fig. 12, v1 to v10 are contour vectors for the outline of the pattern composed of the outer pixels. The contour vector can be derived and stored in the storage by the user clicking the outer pixels in turn. As described above, a contour vector forming a pattern may be derived according to the process of deriving the outline pixels. The contour vectors are stored in the storage as incremental complex vectors of outer pixels.
V2 = 1-i, v3 = -i, v4 = -i, v5 = -1-i, v6 = -1, v7 = 1 + i, v8 = i, v2 = v9 = i, v10 = 1 + i.
Since the contour vectors are incremental complex vectors of the outer pixels, the same contour vectors are generated and stored in the storage even when the user changes the starting pixel when clicking the outer pixel.
When the user clicks the pixel P5 and clicks on the pixel P6, the first contour vector v1 is generated as a real number and an imaginary number at the abscissa increment and the ordinate increment of the pixel, respectively, and stored in the storage unit . A plurality of contour vectors for the surrounding pixels constituting the closed circuit are generated and stored in the storage unit.
The contour vector for the outer pixels forming the glyph is derived from the comparison and contrast stoe images and stored in the storage.
The autocorrelation function value (ACF), the norm function value, and the normal scalar value (R) are calculated from the contour vectors when the contour vectors are derived for the patterns included in the comparative grouped image and the contrasted grouped image (ST-141 , A first calculation step).
The autocorrelation function value (ACF) is a shift scalar sum of the incremental complex vectors for the outer pixels forming each glyph, and is calculated by the following equation (1) and stored in the storage unit.
In Equation (1), v is an incremental complex vector of the outline pixel, and k is the number of outline pixels constituting a certain pattern.
Below
The autocorrelation function value (ACF) is calculated and stored in the storage unit for the patterns included in the comparative grouped image and the contrasted grouped image. Likewise, the auto correlation function (ACF) is calculated for the patterns included in the contrast group images and stored in the storage unit as the pattern DB of the contrast group.
The norm function value for each glyph from the contour vectors of the outline pixels constituting the patterns included in the comparative group image and the contrast group image is calculated by the following
In Equation (2), v is the incremental complex vector of the outline pixel, and k is the number of outline pixels forming the pattern.
The normalized scalar product is calculated from the autocorrelation function value ACF and the norm function value calculated as described above and stored in the storage unit. The normal scalar value (R) of the pattern of the plurality of comparative grouping images for the pattern of the comparative grouping image is calculated as shown in Equation (3) below and stored in the storage unit.
In Equation (3), n is the number of outline pixels constituting the pattern included in the comparative grouping image and the number of outline pixels forming the pattern included in the contrasting grouping image.
In the comparative group image, when there are three patterns having closed pixels constituting a closed circuit, and a plurality of contrast group images stored in the storage unit include three patterns having outer pixels constituting a closed circuit in the contrast grouped images, The normal scalar values R of the nine patterns included in the contrast pattern image are calculated as shown in
Before the autocorrelation function value, the norm function value, and the class correlation function value to be described below are calculated, the number of outline pixels for each pattern included in the comparison pattern image and the contrast pattern image included in the comparison group image is counted, Lt; / RTI > The autocorrelation function value, the norm function value and the class correlation function value are calculated for each pattern included in the contrast pattern images having the same number of pixels and the number of pixels included in the comparative group image.
The normal scalar value R calculated according to the above procedure is displayed on the display together with the comparative strain image. When the normal scalar value (R) is larger than 0.96, the patterns included in the comparative and the comparative stiffness images are found to be the same or similar.
Thus, only when the normal scalar value R is greater than the normal scalar set value (e.g., 0.96) stored in the storage unit, the shoe DB for the contrast stitch is displayed on the display together with the normal scalar value R desirable.
Next, the class correlation function value (ICF) is calculated from the contour vectors derived for the patterns included in the comparison group image and the contrast group image (ST-143, second calculation step).
(ICF) for each pattern included in the contrast streak image and the pattern included in the comparison streak image is calculated by the following equation (4).
Where v is a contour vector of the outline pixel of the pattern included in the comparative grouping image, v 'is a contour vector of the contour pixel of the pattern included in the contour image, and k is the number of contour pixels.
The class correlation function value (ICF) is calculated and stored in the storage unit, and displayed on the display unit as the pedestal DB of each contrast zone together with the comparison frame image. It is confirmed that the similarity of both images is larger as the correlation coefficient value (ICF) is larger.
It is preferable that when the grade correlation function value ICF is displayed on the display unit as the pedestrian DB of each contrast streak, the grade correlation function value ICF is displayed sequentially from a large value.
As a third calculation step (ST-145), the relative distance of the pattern included in the footprint image is calculated.
The shoe DB includes data of the relative footstep origin position of the contrast streak image calculated from the contrast streak images, contrast center data of the center of each pattern included in the contrast streak images, the contrast streak origin position, And further includes contrast speckle gauging distance data, which is the distance between the pattern center positions.
The method for providing the stigma retrieval data further includes a step of computing the relative distance of each pattern included in the comparative group image, and a step of providing distance similarity data;
Wherein the step of computing the comparative core stiffness includes calculating the origin position (comparison root image origin position data) of the comparative rootstock image in the comparative rootstock image, calculating the central position of each pattern included in the comparative rootstock image Calculating a distance between an origin position of the comparison group image and each center position of the comparison group (comparative arch span distance data);
The distance similarity data providing step may include a step of calculating a distance similarity degree of each pattern included in the contrast pattern images for each pattern included in the comparison group image and a step of calculating maximum distance similarity values for each pattern included in the comparison group image And calculating a mean distance similarity value for the derived maximum distance similarity values, and the calculated result is stored in the storage unit. The display unit displays the average distance similarity of the respective contrast shakes to the comparative footsteps.
The distance similarity (tn, tn ') for each pattern included in the contrast image for each pattern included in the comparative group image is calculated by the following equation (5).
Where dn is the distance of each pattern contained in the comparative footprint image, and dn 'is the pattern distance of each pattern contained in the contrast footprint images.
The origin of the footstep can be the center of the pixel that makes up each footstep. In addition, the value obtained by dividing the maximum value and the minimum value of the horizontal coordinate of the pixel forming the footprint image by 2 is the abscissa of the origin, and the value obtained by dividing the sum of the maximum value and the minimum value of the ordinate by 2 is the ordinate of the origin have.
The center of each glyph can be the center of a pixel that forms a pattern. The value obtained by dividing the maximum value and the minimum value of the abscissa of the pixel forming the pattern as described above and dividing by 2 becomes the abscissa of the center of the glyph, and the value obtained by dividing the sum of the maximum value and the minimum value of the ordinate by 2 is the ordinate .
The distance between the center of the pattern and the origin is calculated by adding the square of the difference in the abscissa value to the square of the difference in the ordinate value and the square root thereof.
The similarity of the patterns of the respective patterns included in the plurality of contrast group images is calculated with respect to the pattern distances of the respective patterns included in the comparative group image by the same operation as in Equation (5).
In the case of dn '> 2dn in Equation (5), it is calculated to be 0.
As a fourth calculation step (ST-147), the relative angle of the pattern included in the footprint image is calculated.
The shoe DB stores the origin position data of the contrast shoe images calculated in the contrast shoe images, the contrast center shoe center data as the center positions of the respective patterns included in the contrast shoe image, and the contrast shoe angle data .
The method for providing a footprint search data further includes a step of computing a relative footstep angle and a step of providing an angle similarity data, in which a pattern angle for each pattern included in the comparative footprint image is calculated.
The step of computing the comparative core stiffness angle includes a step of calculating an origin position (comparison root image origin position data) of the comparative grouping image in the comparative grouping image, a step of calculating a center position of each pattern included in the comparative grouping image ), And a central angle of each pattern relative to the origin position of the comparison group image (comparison pattern image angle data) is calculated.
The step of providing the degree of similarity data may include the steps of calculating the degree of similarity of each pattern included in the contrasted grouped images for each pattern included in the comparative grouped image, and calculating the maximum degree of similarity values for each pattern included in the comparative grouped image And calculating an average degree of similarity value for the derived maximum degree of similarity values.
The display unit displays the average degree of similarity of each contrasting rock to the comparison streak.
The angular similarities (? N,? N ') of the respective patterns included in the contrast image for each pattern included in the comparative group image are calculated as shown in Equation (6).
In the above equation, rn is the pattern angle of each pattern included in the comparative group image, and rn 'is the pattern angle of each pattern included in the contrast pattern image. In the above, the angle can be calculated by using a line extending the maximum value and the minimum value of the ordinate as a reference line.
In the case of rn '> 2rn in Equation (6), 0 is calculated.
The method of providing the streak search data includes the steps of calculating a relative angle of each of the patterns included in the comparative image of the striae, An operation step, and a similarity data providing step;
The step of computing the comparative core stiffness angle includes a step of calculating an origin position (comparison root image origin position data) of the comparative grouping image in the comparative grouping image, a step of calculating a center position of each pattern included in the comparative grouping image And calculating a center angle of each pattern relative to an origin position of the comparison group image (comparison pattern image data).
Wherein the step of computing the comparative core stiffness comprises the step of calculating a distance between the origin position of the comparative grouping image and the center position of each pattern (comparative legend image pattern distance data);
Wherein the similarity data providing step includes the steps of calculating distance similarities of the patterns included in the contrasting stone images for the respective patterns included in the comparison group image,
Calculating the degree of similarity of each of the patterns contained in the contrasted root image for each pattern included in the comparative group image,
Calculating a degree of similarity of each pattern included in the contrasted root image for each pattern included in the comparison root image from the distance similarity and the degree of similarity;
Extracting maximum similarity values for each pattern included in the comparative group image, and calculating an average similarity value for the derived maximum similarity values.
The display unit displays the average degree of similarity of each contrasting shake with respect to the comparative grouping.
The angle similarity (? N,? N ') for each pattern included in the contrast image for each pattern included in the comparison group image is calculated by Equation (6)
The distance similarity (tn, tn ') for each pattern included in the contrast image of each pattern included in the comparative group image is calculated by the above equation (5)
The similarity (Sn, Sn ') of each pattern included in the contrast image for each pattern included in the comparative group image is calculated as shown in Equation (7).
10: footprint image 11: first feature
13: second characteristic portion 15: third characteristic portion
Claims (8)
Calculating a contour vector for outline pixels of at least one of the glyphs included in the comparison group image stored in the storage unit; calculating an autocorrelation function value for each glyph from the contour vector; Calculating a norm function value for each of the patterns by comparing an autocorrelation function value of each contrast stance with an autocorrelation function value and a norm function value of the comparison group and a normal scalar value R of a norm function value, And displaying a normal scalar value of each contrast zone on the display; Wherein the outline is a closed loop, and the contour vector of each pattern is an incremental complex vector vector of outline pixels constituting the pattern.
The function value (AC ACF)) for each pattern of the contrast streak or the comparison streak is the scalar sum of the incremental complex vector for the outer vector forming each glyph
The normal color value (R)
Where ACF (n) is a shift scalar sum computed for the glyphs contained in the comparison root image, ACF '(n) is the contour vector for the contour pixel, (N) is the computed function value for the pattern included in the comparative grouping image, and | ACF '(n) | is the contrast value of the contrasting sketch image Is a genomic function value calculated for a glyph included in the segment search data.
In the above, the class correlation function value (ICF) for each pattern of contrast streaks and each pattern of comparative group shakes
In the above equations, v is a contour vector for the outline pixels of each pattern included in the comparative group image, v 'is a contour vector for the contour pixels of each pattern included in the contrast pattern image, and k is the number of contour pixels being.
The method for providing the stigma retrieval data further includes a step of computing the relative distance of each pattern included in the comparative group image, and a step of providing distance similarity data;
Wherein the step of computing the comparative core stiffness includes calculating the origin position (comparison root image origin position data) of the comparative rootstock image in the comparative rootstock image, calculating the central position of each pattern included in the comparative rootstock image Calculating a distance between an origin position of the comparison group image and each center position of the comparison group (comparative arch span distance data);
The distance similarity data providing step may include a step of calculating a distance similarity degree of each pattern included in the contrast pattern images for each pattern included in the comparison group image and a step of calculating maximum distance similarity values for each pattern included in the comparison group image And calculating an average distance similarity value for the derived maximum distance similarity values;
Wherein the display unit displays the average distance similarity of each of the contrast shakes relative to the comparison shoe.
The distance similarity (tn, tn ') for each pattern included in the comparative footprint images for each pattern included in the comparative footprint image is
Where dn is the pattern distance of the angle included in the comparative footprint image, and dn 'is the pattern distance of each pattern included in the contrast footprint images.
Wherein the method for providing the footstep search data further includes a step of calculating a relative footstep angle and a step of providing the degree of similarity data, the flint angle being calculated for each pattern included in the comparative footstep image;
The step of computing the comparative core stiffness angle includes a step of calculating an origin position (comparison root image origin position data) of the comparative grouping image in the comparative grouping image, a step of calculating a center position of each pattern included in the comparative grouping image Calculating a center angle of each of the patterns relative to an origin position of the comparative group image (comparison pattern image angle data);
The step of providing the degree of similarity data may include the steps of calculating the degree of similarity of each pattern included in the contrasted grouped images for each pattern included in the comparative grouped image, and calculating the maximum degree of similarity values for each pattern included in the comparative grouped image And calculating an average degree of similarity value for the derived maximum degree of similarity values;
Wherein the display unit displays the average degree of similarity of each of the contrast shakes relative to the comparison shoe.
The angle similarity (? N,? N ') of each pattern included in the contrast pattern images for each pattern included in the comparative group image
In the above equation, rn is the pattern angle of each pattern included in the comparative group image, and rn 'is the pattern angle of each pattern included in the contrast pattern image.
The method for providing the stigma retrieval data includes a step of computing the relative angle of each of the patterns included in the comparative group image, a step of computing the relative distance of each pattern included in the comparative group image, And a similarity data providing step;
The step of computing the comparative core stiffness angle includes a step of calculating an origin position (comparison root image origin position data) of the comparative grouping image in the comparative grouping image, a step of calculating a center position of each pattern included in the comparative grouping image Calculating a center angle of each of the patterns relative to an origin position of the comparative group image (comparison pattern image angle data);
Wherein the step of computing the comparative core stiffness comprises the step of calculating a distance between the origin position of the comparative grouping image and the center position of each pattern (comparative legend image pattern distance data);
Wherein the similarity data providing step includes the steps of calculating distance similarities of the patterns included in the contrasting stone images for the respective patterns included in the comparison group image,
Calculating the degree of similarity of each of the patterns contained in the contrasted root image for each pattern included in the comparative group image,
Calculating a degree of similarity of each pattern included in the contrasted root image for each pattern included in the comparison root image from the distance similarity and the degree of similarity;
Calculating maximum similarity values for each pattern included in the comparison group image, and calculating an average similarity value for the derived maximum similarity values;
Wherein the display unit displays the average similarity of each of the contrast shakes to the comparison shoe.
The angle similarity (? N,? N ') for each pattern included in the contrast pattern images for each pattern included in the comparative group image
Where rn is the gyration angle of the angle included in the comparative group image, and rn 'is the gyration angle of each glyph included in the contrast group image;
The distance similarity (tn, tn ') for each pattern included in the contrast image for each pattern included in the comparative group image is
Where dn is the gyration distance of the angle included in the comparative group image, dn 'is the gyration distance of each glyph included in the contrast group image,
The similarity (Sn, Sn ') of each pattern included in the contrast image for each pattern included in the comparative group image
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020160016573A KR101781359B1 (en) | 2016-02-12 | 2016-02-12 | A Method Of Providing For Searching Footprint And The System Practiced The Method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020160016573A KR101781359B1 (en) | 2016-02-12 | 2016-02-12 | A Method Of Providing For Searching Footprint And The System Practiced The Method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR20170095062A true KR20170095062A (en) | 2017-08-22 |
| KR101781359B1 KR101781359B1 (en) | 2017-09-26 |
Family
ID=59757922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020160016573A KR101781359B1 (en) | 2016-02-12 | 2016-02-12 | A Method Of Providing For Searching Footprint And The System Practiced The Method |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR101781359B1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257662A (en) * | 2020-11-12 | 2021-01-22 | 安徽大学 | Pressure footprint image retrieval system based on deep learning |
| CN112800267A (en) * | 2021-02-03 | 2021-05-14 | 大连海事大学 | Fine-grained shoe print image retrieval method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005027951A (en) | 2003-07-08 | 2005-02-03 | Gen Tec:Kk | Sole database and creating method thereof |
| JP2007226756A (en) | 2006-02-22 | 2007-09-06 | Univ Kinki | Method and device for evaluating left image such as footprint using ridgelet transform |
-
2016
- 2016-02-12 KR KR1020160016573A patent/KR101781359B1/en active IP Right Grant
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257662A (en) * | 2020-11-12 | 2021-01-22 | 安徽大学 | Pressure footprint image retrieval system based on deep learning |
| CN112800267A (en) * | 2021-02-03 | 2021-05-14 | 大连海事大学 | Fine-grained shoe print image retrieval method |
| CN112800267B (en) * | 2021-02-03 | 2024-06-11 | 大连海事大学 | Fine-granularity shoe print image retrieval method |
Also Published As
| Publication number | Publication date |
|---|---|
| KR101781359B1 (en) | 2017-09-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6091560B2 (en) | Image analysis method | |
| CN110738101A (en) | Behavior recognition method and device and computer readable storage medium | |
| Lin et al. | A multi-level morphological active contour algorithm for delineating tree crowns in mountainous forest | |
| US20050053276A1 (en) | Method of obtaining a depth map from a digital image | |
| JP6648925B2 (en) | Image processing method, image processing device, image processing system, production device, program, and recording medium | |
| KR102073468B1 (en) | System and method for scoring color candidate poses against a color image in a vision system | |
| Goodbody et al. | Digital aerial photogrammetry for assessing cumulative spruce budworm defoliation and enhancing forest inventories at a landscape-level | |
| JP2006285310A (en) | Evaluation method of canopy of forest, and its canopy evaluation program | |
| US9292929B2 (en) | Image region extraction device, image region extraction method, and image region extraction program | |
| KR101510206B1 (en) | Urban Change Detection Method Using the Aerial Hyper Spectral images for Digital Map modify Drawing | |
| US11170215B1 (en) | System and method for discriminating and demarcating targets of interest in a physical scene | |
| US11049268B2 (en) | Superimposing position correction device and superimposing position correction method | |
| JP2013030183A (en) | Environment recognition device, and program | |
| KR101781359B1 (en) | A Method Of Providing For Searching Footprint And The System Practiced The Method | |
| EP1760636B1 (en) | Ridge direction extraction device, ridge direction extraction method, ridge direction extraction program | |
| JP2017003525A (en) | Three-dimensional measuring device | |
| JP4721829B2 (en) | Image retrieval method and apparatus | |
| JP6293505B2 (en) | Tool inspection method and tool inspection apparatus | |
| CN109657540B (en) | Withered tree positioning method and system | |
| US11562505B2 (en) | System and method for representing and displaying color accuracy in pattern matching by a vision system | |
| JP3919722B2 (en) | Skin shape measuring method and skin shape measuring apparatus | |
| JP2014146181A (en) | Operation information recording device, display system, and program | |
| CN110427961B (en) | Building information extraction method and system based on rule and sample fusion | |
| CN109034138B (en) | Image processing method and device | |
| KR100954137B1 (en) | Edge-based text localization and segmentation algorithms for automatic slab information recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E902 | Notification of reason for refusal | ||
| E701 | Decision to grant or registration of patent right | ||
| GRNT | Written decision to grant |