JP2008107228A - Route measuring apparatus, route measuring method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a route measuring method and a route measuring apparatus, capable of speedily and precisely measuring the minimum distance between objects, even in the case that no candidate of a route connecting the objects is given explicitly, and an obstruction through which the route can not pass, is exists between the objects to be measured being indicated on an image. <P>SOLUTION: A cost field which prescribes the ease of passing of the route, is generated by a cost field generating apparatus 4, based on a locational relation between a distance measurement object and a non-object through which the route can not pass. A plurality of routes connecting the objects are generated dynamically by a dynamic route generating apparatus 5 while referring to values of the cost field at respective image points. Then, the shortest route connecting the objects is searched by a route searching apparatus 6. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a route measuring device and a route measuring method, and more particularly to a route measuring device and a route measuring method for measuring a route between objects displayed on an image.

  Conventionally, as a method or apparatus for measuring the distance between objects displayed on an image, there is a “medical image apparatus” disclosed in, for example, Japanese Patent Laid-Open No. 2005-65845 (Patent Document 1).

  In Patent Document 1, one point (hereinafter referred to as a point A) of two points designated on a medical image by an operator is expanded by the region expansion method, and the increment of the region is the other point (hereinafter referred to as a point B). Repeat the expansion until it reaches In the expansion process, the relationship between the parent pixel and the child pixel is recorded, and when the expanded area reaches the point B, the relationship between the parent pixel and the child pixel is traced back to connect the point A and the point B. Measure the length of the path.

  However, in the method disclosed in Patent Document 1, if there is an obstacle that cannot be passed between two points whose distance is to be measured, the route obtained by repeating the region expansion method is not necessarily the shortest route. There is a problem. In particular, the shortest path extending along the boundary of the obstacle cannot be obtained by the method shown in Patent Document 1.

  In the medical field, the shortest distance between objects is often important information, such as the penetration rate of the administered drug or the presence or absence of interaction between structures in the cell nucleus, and the shortest distance is not always obtained. When the above-described conventional method is applied, there is a problem that accurate information cannot be obtained.

As a related technique, Japanese Patent Laid-Open No. 2003-303344 (Patent Document 2) discloses an image region extraction method and apparatus.
In this prior art, the image data space is coarse-grained, the coarse-grained empirical probability distribution is calculated, the parameters are initialized, the coarse-grained conditional probability distribution is calculated, the class membership probability is calculated, and the parameters are updated. Then, the evaluation function is calculated, and these processes are repeated until there is no change in the evaluation function. When there is no change in the evaluation function, an image region is extracted based on the class membership probability.

Japanese Patent Laid-Open No. 2006-133012 (Patent Document 3) discloses a route search device and a route search data generation method.
In this prior art, a route search device that searches for and outputs an optimum route to a destination is a geometrical combination of point data having coordinates, line data having a point sequence, and surface data having a point sequence that goes around the boundary. Route search data consisting of a combination of point / line / surface mixed data including graphic data and line data that is combined with geometric graphic data to set a representative point in the surface data to form a network in both directions, and route search Route search means for performing a route search based on the data for use, and a route output means for drawing and outputting the route searched by the route search means by the point / line / plane mixed method based on the geometric figure data. Features.

Japanese Patent Laid-Open No. 5-46590 (Patent Document 4) discloses a graph shortest path search method and apparatus for obtaining a plurality of optimum solutions.
In this prior art, as a method for obtaining the shortest path of a graph, first, the shortest cost between an arbitrary vertex of a directed graph with a real value cost attached to the edge between vertices and the end point t is calculated, and then the directed graph G The sub-path from the starting point s is subordinate to the sum f (Xi) of the accumulated cost g (Xi) from the starting point s and the shortest cost h (Xi) of the remaining paths according to the set analysis conditions. While pruning, the route is sequentially expanded toward the end point t like a and a + 1, and the top N optimum routes held when the search is completed are output.

Japanese Patent Laid-Open No. 11-184837 (Patent Document 5) discloses a shortest path search system.
In this prior art, a function called an evaluation value h (v) is used as an index for reducing the route to be searched. The evaluation value h (v) is defined as the lower limit value of the path length from the current point v to the end point T. As a method for obtaining this lower limit value, pseudo code is used. According to this pseudo code, in order to obtain the lower limit value, it is necessary to follow the path in the reverse direction from the end point T toward the point included in the set U of points for which the lower limit value is desired. That is, in the prior art of Patent Document 5, a plurality of routes connecting the start point to the end point must be set in advance.

Furthermore, a wiring design apparatus is disclosed in Japanese Patent No. 2943358 (Patent Document 6).
In this prior art, the region is expanded from the start point, and when the boundary of the region reaches the end point, the route is found by tracing back from the end point to the start point. At this time, there is a step of generating a route. However, the generated route is not a plurality but one, and therefore there is no need to search for a route. In addition, it is not always the shortest path in terms of space.

JP 2005-65845 A JP 2003-303344 A JP 2006-133012 A Japanese Patent Laid-Open No. 5-46590 JP 11-184837 A Japanese Patent No. 2943358

  An object of the present invention is to provide a path measurement device and path measurement capable of measuring the shortest distance between objects at high speed and with high accuracy even when an obstacle exists between the measurement objects displayed on the image. It is to provide a method.

  In the following, means for solving the problem will be described using the numbers used in [Best Mode for Carrying Out the Invention] in parentheses. These numbers are added to clarify the correspondence between the description of [Claims] and [Best Mode for Carrying Out the Invention]. However, these numbers should not be used to interpret the technical scope of the invention described in [Claims].

  The route measuring device of the present invention is a route measuring device that measures a route between objects input from an image with respect to an object for which a candidate for a route is not determined in advance. The path measuring device of the present invention includes an image input device (1) that inputs an object of distance measurement and an image of a non-object that is excluded from the path of distance measurement, and an object and a non-object included in the input image. Image binarization device (2) that separates the inside and outside of the object area by binarization, a start point that is the start point of the route, a destination point that is the end point of the route, a non-object region that the route cannot pass through, Based on the parameter setting device (3) for setting the path expansion parameter r and the start point, the target point, and the non-object region, the closer to the start point and the farther from the target point, the larger the value, and By taking a large value inside the non-object area, a cost field that defines the cost that defines the ease of passage of the route at each point of the input image and generates a cost field that is a set of points for which the cost is defined Generator (4), open A starting point set having a starting point as a single element is generated starting from the point, a line segment set of line segments of length r extending from each starting point included in the starting point set is generated, and a set of end points of the line segment The existing start point set is updated as a new start point set, and the generation and update of the start point set and the line segment set are sequentially repeated until all the paths represented by the line set reach the target point. Dynamic route generation device (5) to be generated, route search device (6) for searching for the shortest route from routes included in a set of routes, and a route display device for displaying the length of the shortest route and the shortest route (7).

  After obtaining the shortest path, the path measuring device of the present invention checks whether the length r of the line segment satisfies a given condition, and reduces the length of the line segment if the condition is not satisfied, A route reconstructing device (8) is further provided for holding the vicinity of the shortest route as a non-object in the binary image.

  In the process of generating a route, the dynamic route generation device (5) is configured such that the distance between the end point of a line segment constituting one route and the end point of a line segment constituting another route is smaller than a given value. If this happens, these end points are merged into one.

  The route search device (6) determines a route by the Viterbi algorithm.

  The route search device (6) determines a route by the Dijkstra algorithm.

  The route measurement method of the present invention is a route measurement method for measuring a route between objects for which a route candidate is not determined in advance. And the path | route measuring method of this invention is the 1st process which inputs the image of the object of distance measurement, and the non-object excluded from the path | route of distance measurement, and the object and non-object included in an input image. A second step of separating the inside and the outside of the area by binarization, a start point that is the start point of the path, a destination point that is the end point of the path, a non-object area through which the path cannot pass, and an extension parameter r of the path Based on the third step of setting and the start point, the target point, and the non-object region, the closer to the start point and the farther from the target point, the larger the value and the larger value inside the non-object region By defining the cost that defines the ease of passage of the route at each point of the input image, the fourth step of generating a cost field that is a set of points for which the cost is defined, and the starting point The starting point is just one element Generating a start point set, generating a line segment set consisting of line segments of length r extending from each start point included in the start point set, updating the existing start point set with the end point set of line segments as a new start point set, Included in the route set is the fifth step of generating the set of routes by sequentially repeating the generation and update of the start point set and line segment set until all the routes represented by the line segment set reach the destination point. A sixth step of searching for the shortest route from among the routes and a seventh step of displaying the shortest route and the length of the shortest route are included.

  The path measurement method of the present invention, after obtaining the shortest path, checks whether the length r of the line segment satisfies a given condition, and reduces the length of the line segment if the condition is not satisfied, An eighth step of holding the vicinity of the shortest path as a non-object in the binary image is further included. Further, the third, fourth, fifth and sixth steps are repeated until a given condition is satisfied.

  In the route measurement method of the present invention, in the fifth step, in the process of generating a route, the distance between the end point of a line segment constituting one route and the end point of a line segment constituting another route is given. When it becomes smaller than the value, these end points are integrated into one.

  In the route measuring method of the present invention, the route is determined by the Viterbi algorithm in the sixth step.

  In the route measurement method of the present invention, the route is determined by the Dijkstra algorithm in the sixth step.

  The program of the present invention causes a computer to execute any one of the above-described path measurement methods.

  Based on the arrangement of the target and non-target objects of the path measurement, a cost field that defines the ease of the path at each point in the image is generated, and the shortest path candidate is dynamically generated while referring to the cost at each point. As a result, it is possible to perform path measurement at a higher speed and with higher accuracy than the conventional method.

  A first embodiment of the present invention will be described below with reference to the accompanying drawings. In the present embodiment, a case will be described as an example in which a microscope image of a stained cell is used as an input and the distance between cell tissues (chromosome, cell nucleus, nucleolus, etc.) is measured. The cell image in this case may be data in which cell nuclei, chromosomes, nucleoli, etc. are recorded in separate monochrome images. If the areas of each object do not overlap, a plurality of objects and non-objects may be combined. It may be displayed on two monochrome images. Alternatively, it may be a single color image in which cell nuclei, chromosomes, nucleoli, etc. are recorded by different color channels. In other cases, if the data is recorded as an image, it can be executed in the same procedure as the method described below.

  Hereinafter, the route measuring apparatus of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing the configuration of the path measuring apparatus of the present invention.

  Referring to FIG. 1, a path measuring apparatus according to the present invention includes an image input apparatus 1, an image binarizing apparatus 2, a parameter setting apparatus 3, a cost field generating apparatus 4, a dynamic path generating apparatus 5, and a path. A search device 6 and a route display device 7 are provided.

  The image input device 1 reads image data. The image binarization apparatus 2 binarizes the image. The parameter setting device 3 specifies an object and a non-object in the image. The cost field generation device 4 generates a cost field from the arrangement of objects and non-objects. The dynamic path generation device 5 dynamically generates a path connecting the objects. The route search device 6 searches for a route from a plurality of generated routes. The route display device 7 displays the final route and the shortest distance between objects.

2A and 2B are flowcharts showing the processing procedure of the route measuring method of the present invention.
(1) Step S1
The input device 1 inputs an image of a cell taken with a biological microscope or the like. The input device 1 sends the read data to the image binarization device 2.
(2) Step S2
The image binarization apparatus 2 binarizes the read image of each object based on the luminance value of each point on the image. When binarization is performed, for example, it is possible to set a luminance value larger than a given threshold value to 255 and a luminance value less than the threshold value to 0. In addition, it is also possible to normalize the sum of luminance values on the image to 1 and to binarize stochastically considering the image as a two-dimensional random field. The binarization using the random field is described in detail in, for example, Japanese Patent Laid-Open No. 2003-303344 (Patent Document 2). The image binarization apparatus 2 sends the input image and the binarized image to the parameter selection apparatus 3.

  An example of a binarized image for the chromosome 11 and the nucleolus 12 is shown in FIG. In FIG. 3, a large cluster at the center extracts the nucleolus 12 and displays the region in white, and two clusters around it are the chromosome 11. When the chromosomes interact with each other, the nucleolus 12 becomes an obstacle. Therefore, to measure the distance between the two chromosomes, it is necessary to bypass the nucleolus. In this case, the two chromosomes 11 are objects, and the nucleolus 12 is a non-object.

(3) Step S3
The parameter setting device 3 sets a start point S for performing path measurement, a target point O, a non-object area through which the path cannot pass, and a path expansion parameter r. To specify the start point S and the target point O, for example, an image in which only an object including the start point S is recorded, an image in which only an object including the target point O is recorded, and only a non-object is recorded. After the read images are sequentially read and binarized by the image binarization apparatus 2, the center of gravity of each region can be used as the start point and the target point. As the non-object, a binarized image obtained by recording only the non-object is used as an area as it is. In addition to this, for example, it is also possible for the operator to input the coordinate value through an external input device. The parameter setting device 3 sends the input image, the binarized image, the start point S, the target point O, the non-object region represented by the binary image, and the expansion parameter r to the cost field generating device 4.
(4) Step S4
The cost field generating device 4 quantifies the ease of passing the route at each point on the image based on the arrangement of the target and non-target objects of the route measurement. In the following, the ease of passing a digitized route is called a cost, and the entire cost defined by each point is called a cost field. In order to define the cost at the point x on the image, for example, the following function is used.

Here, x _s and x _o are the coordinates of the start point S and the target point O, respectively. a and b are arbitrary positive real constants, a controls the strength with which the path is made from the starting point, and b controls the strength with which the path is attracted to the destination point. In order to reflect non-objects that cannot be routed in the cost field, for example, the following function is used.

  Here, I (x) is an instruction function that takes 1 when the point x is included in the non-object and 0 otherwise. Further, c is an arbitrary positive real constant, but it is desirable to take a large value so that the path cannot pass through the non-object region.

  The function that defines the cost field is not limited to the one described above, and takes a larger value as it is closer to the starting point S and further away from the target point O, and takes a larger value within the object region. I just need it.

  An example of the cost field is shown in FIG. In FIG. 4, S represents the start point and O represents the target point. FIG. 4A shows the cost value at each point in shades, where the lighter the shade, the greater the cost, and the darker the cost. FIG. 4B is a three-dimensional display of the same cost field. By using such a cost field, efficient route generation becomes possible.

  The cost field generation device 4 sends the generated cost field, the start point S, the target point O, and the non-object region represented by the binary image to the dynamic path generation device 5. The dynamic route generation device 5 dynamically generates a plurality of routes from the start point S to the destination point O.

  In the following description, a route is assumed to be composed of a plurality of line segments, and each line segment is a straight line having a length r connecting a start point and an end point. The start point (or end point) of each line segment constituting the path shares the end point (or start point) with at least one other line segment, and thus at least one path connects the start point S and the destination point O. This is assured from the path configuration method described below.

(5) Step S5
First, the dynamic route generation device 5 initializes a counter k for sequential update of routes to zero.
(6) Step S6
Next, the dynamic route generation device 5 initializes the route set P _k with an empty set, and the start point set I _k with a set consisting of only one point of the start point S.
(7) Step S7
Next, the dynamic route generation device 5 initializes a set E _k of line segments constituting a part of the route with an empty set.
(8) Step S8
Next, the dynamic route generation device 5 selects one start point to which no end point is assigned from the start point set I _k . This is represented by v.
(9) Step S9
Next, the dynamic route generation device 5 checks whether v is within the distance r from the target point O.
(10) Step S10
If v is further away from the target point O than the distance r, the dynamic path generator 5 first selects N points that are separated from the distance r by v (N is given). In order to select such a point, for example, each point obtained by dividing the circumference of the radius r centering on v into N may be used. N line segments are generated, each having N points selected with v as the starting point.
(11) Step S11
If v is within r from the target point O, the dynamic path generation device 5 generates a line segment v-O with the target point O as an end point, and adds the generated line segment to the line segment set E _k . To do.
(12) Step S12
Next, the dynamic path generator 5 adds the N line segments generated in step S10, or generated in step S11, one line segment extending target point O to the line set E _k from v .
(13) Step S13
Next, the dynamic path generation device 5 examines whether or not the end point is given to all the points belonging to I _k and a line segment is generated. If there is a start point to which no end point is given, step S8 and subsequent steps are repeated.
(14) Step S14
If end points are assigned to all the start points belonging to I _k , the dynamic path generation device 5 checks whether all the end points match the destination point O. If all the end points coincide with the target point O, one or more paths connecting the start point S to the target point O have been generated, so the dynamic path generation device 5 uses the path set P _k , the cost. The field, the start point S, the target point O, and the non-object region represented by the binary image are sent to the route search device 6.
(15) Step S15
If there is a start point given an end point different from the target point O, the dynamic path generation device 5 first calculates the cost value of the point through which the line segment passes for each line segment belonging to the line segment set E _k . The sum is calculated, and all line segments having a value equal to or higher than the given cost value C are collected, and this set is _defined as D _k .
(16) Step S16
Next, the dynamic path generator 5, among line segments belonging to the E _k, a set of line segments that do not belong to the D _k to (this referred to as E _k -D _k), examining the end point, the end point Let T _k be the set. However, if the end point is consistent with the desired point O are excluded from T _k.
(17) Step S17
Next, the dynamic route generation device 5 sets the end point set T _k as the next sequential update start point set I _{k + 1,} and adds the line segments belonging to E _k -D _k to the currently obtained route set P _k. In addition, let this route set be P _{k + 1} .
(18) Step S18
Next, the dynamic route generation device 5 increments the value of the counter k by 1, and repeats step S7 and subsequent steps.

FIG. 5 shows an example of dynamic route generation from step S5 to step S18. In FIG. 5, S is the start point of one path, v _i (i = 1, 2,..., 6) is the end point of the line segment, and e _i (i = 1, 2,..., 6) is the end of v _i . Represents a line segment. k is the number of sequential updates, and the start point set I, line segment set E, end point set T, deleted line segment D, and generated route set P at each update time are shown together with the figure.

  In the dynamic route generation device 5, in the process of generating a route, the distance between the end point of a line segment constituting one route and the end point of a line segment constituting another route becomes smaller than a given value. In such a case, it is possible to suppress generation of an extra route by integrating these end points into one. This mechanism enables more efficient path generation.

(19) Step S19
The route search device 6 searches for the shortest route from the finally obtained route set _Pk and calculates the shortest distance. For the search for the shortest path, for example, a Viterbi algorithm used for signal processing or the like can be used. In this case, the counter k that sequentially updates the route is regarded as time, and the distance of the line segment may be used as it is as the weight assigned to each line segment that forms the path. Further, if the line segment constituting the route is regarded as an edge and the start point and end point of the line segment are regarded as vertices, the Dijkstra method can be used as a method for searching for the optimum route.
(20) Step S20
The route display device 7 receives and outputs the shortest route and its distance obtained by the route search device 6.

  A second embodiment of the present invention will be described below with reference to the accompanying drawings. In the embodiment described below, the description of the same part as the first embodiment described above is omitted.

  FIG. 6 is a block diagram showing the configuration of the path measuring apparatus according to the second embodiment of the present invention. The route measurement device according to the second embodiment of the present invention includes an image input device 1, an image binarization device 2, a parameter setting device 3, a cost field generation device 4, a dynamic route generation device 5, and a route search. In addition to the device 6 and the route display device 7, a route reconstruction device 8 is further provided.

  The image input device 1, the image binarization device 2, the parameter setting device 3, the cost field generation device 4, the dynamic route generation device 5, the route search device 6, and the route display device 7 are the same as in the first embodiment. is there. The route reconstructing device 8 receives the shortest route and the route extension parameter from the route search device 6, checks whether the value of the route extension parameter is smaller than a given value, and the value of the route extension parameter is less than the given value. Then, the shortest route and the length of the route are sent to the route display device 7. If the value of the path expansion parameter is equal to or greater than a given value, a number obtained by multiplying the path expansion parameter by a given positive real number smaller than 1 is set as a new expansion parameter. Then, an area other than the vicinity of the shortest path is expressed as a new non-object area in the binary image, and an area other than the vicinity of the shortest path expressed by the updated extension parameter and the binary image is sent to the parameter setting device 3. .

7A and 7B are flowcharts showing the processing procedure of the route measuring method according to the second embodiment of the present invention. In addition, about step S1-step S20 in FIG. 7A and FIG. 7B, it is the same as that of 1st Embodiment. The following processing is performed between step S19 and step S20.
(21) Step S21
The route reconstructing device 8 receives the shortest route and the route extension parameter r from the route search device 6 and checks whether the value of the route extension parameter is smaller than a given value r ₀ . If the value of the route expansion parameter is equal to or less than the given value r ₀ , the shortest route and the length of the route are sent to the route display device 7.
(22) Step S22
If the value of the path expansion parameter is equal to or greater than the given value r _0, a number obtained by multiplying the path expansion parameter by a given positive real number ε smaller than 1 is set as a new expansion parameter.
(23) Step S23
Furthermore, a region other than the vicinity of the shortest path is expressed as a new non-object region in the binary image, and the updated extension parameter and the region other than the vicinity of the shortest path expressed in the binary image are sent to the parameter setting device 3. . The vicinity of the shortest path is, for example, the entire area included in the radius d (d is given) around each point constituting the path.

  A smoother path can be obtained by repeating the procedure from step S5 to step S19 described in the first embodiment using the extension parameter and the non-target object region obtained by the path reconstruction device 8. . In addition, by setting a region other than the vicinity of the route obtained first as the non-target object region, the route search region is narrowed down, and more efficient route generation and route search are possible.

The features of the route measuring method and the route measuring device of the present invention will be described in detail below.
The path measurement method according to the first embodiment of the present invention includes the following first, second, third, fourth, fifth, and sixth steps. It has the process and the 7th process, It is characterized by the above-mentioned.
The first step is to input an image of a distance measurement object and a non-object excluded from the distance measurement path in a path measurement method for measuring a path between objects whose path candidates are not determined in advance. . In the second step, the inside and outside of the object and non-object regions included in the input image are separated by binarization. In the third step, a start point that is the start point of the route, a destination point that is the end point of the route, a non-object region through which the route cannot pass, and an extension parameter r of the route are set. The fourth step is based on the start point, the target point, and the non-object region, and takes a larger value closer to the start point and further from the target point, and takes a larger value inside the non-object region. Then, a cost defining the ease of passing the route is defined at each point of the input image, and a cost field which is a set of points where the cost is defined is generated. The fifth step generates a start point set starting from the start point and having only one element as the start point, and generates a line segment set consisting of line segments of length r extending from each start point included in the start point set. Updating the existing start point set with the set of end points of the line segment as a new start point set, and generating the start point set and the line segment set until all paths represented by the line segment set reach the target point, A set of routes is generated by sequentially repeating the update. The sixth step searches for the shortest route from the routes included in the set of routes. In the seventh step, the shortest path and the length of the shortest path are displayed.

  In the path measurement method according to the second embodiment of the present invention, after obtaining the shortest path, it is checked whether or not the length r of the line segment satisfies a given condition, and when the condition is not satisfied, The eighth, third, fourth, fifth, and sixth steps have the eighth step of reducing the length and holding the vicinity of the shortest path as a non-object in the binary image until a given condition is satisfied. It is characterized by repeating.

  In the fifth step, the route measurement method according to the third embodiment of the present invention, in the fifth step, in the process of generating a route, between the end point of a line segment constituting one route and the end point of a line segment constituting another route. When the distance becomes smaller than a given value, these end points are integrated into one.

  The route measuring method according to the fourth embodiment of the present invention is characterized in that the route is determined by the Viterbi algorithm in the sixth step.

  The route measuring method according to the fifth embodiment of the present invention is characterized in that the route is determined by the Dijkstra algorithm in the sixth step.

A path measuring apparatus according to a sixth embodiment of the present invention includes an image input apparatus, an image binarizing apparatus, a parameter setting apparatus, a cost field generating apparatus, a dynamic path generating apparatus, a path searching apparatus, and a path display. And a device.
The image input device inputs an object of distance measurement and an image of a non-object excluded from the distance measurement path. The image binarization device separates the inside and the outside of the object and non-object regions included in the input image by binarization. The parameter setting device sets a start point that is a start point of a route, a destination point that is an end point of the route, a non-object region through which the route cannot pass, and an expansion parameter r of the route. Based on the start point, the target point, and the non-object region, the cost field generation device takes a larger value as it is closer to the start point and further away from the target point, and takes a larger value inside the non-object region. Then, a cost defining the ease of passing the route is defined at each point of the input image, and a cost field which is a set of points where the cost is defined is generated. The dynamic path generation device generates a start point set starting from the start point and having only one element as the start point, and generates a line segment set consisting of line segments of length r extending from each start point included in the start point set. Generate and update the existing start point set using the end point set of the line segment as a new start point set, and generate the start point set and the line segment set until all the paths represented by the line set reach the destination point A set of paths is generated by sequentially repeating the update. The route search device searches for the shortest route from the routes included in the set of routes. The route display device displays the shortest route and the length of the shortest route.

  The path measurement device according to the seventh embodiment of the present invention determines whether the length r of the line segment satisfies a given condition after obtaining the shortest path, and if the condition is not satisfied, It is characterized by having a path reconstruction device that reduces the length and holds the vicinity of the shortest path as a non-object in a binary image.

That is, in the route measurement method of the present invention, the cost field is defined around the object to be measured by the sixth step of the first embodiment of the present invention, and the cost value at each point is referenced. Dynamically generate a route. FIG. 8 shows a line segment 13 (an arrow represented by practice) that becomes a new extension part of the route and a line segment 14 (an arrow represented by a dotted line) that has been deleted to reflect the cost value. In FIG. 8, the whiter the background, the higher the cost, and the darker the cost. A circular white area (non-object area 15) indicates an area where there is a non-object that the path cannot pass. The cost field takes a large value as it approaches the starting point of the route and moves away from the target point by the fourth step of the first embodiment of the present invention, and takes a very large value inside the non-object region. Has been generated. For this reason, the route is generated so as to bypass the area where the non-object exists and to efficiently extend from the start point to the target point. In the present invention, this mechanism makes it possible to generate an efficient route without generating a useless route.
Further, the route search mechanism of the fourth and fifth embodiments of the present invention searches for a route having the shortest distance from a plurality of routes connecting the start point and the target point. Since it is excluded, high-speed path measurement is possible.
Further, as described in the second embodiment of the present invention, first, a route expansion parameter r is set to a large value to roughly obtain the shortest route, the search region is narrowed in the vicinity of the obtained shortest route, and then the reduced expansion is performed. By repeating the route generation using the parameters, it is possible to search for a more accurate and smooth route.

FIG. 1 is a block diagram showing a configuration of a distance measuring device in the first embodiment of the present invention. FIG. 2A is a flowchart showing a processing procedure in the first embodiment of the present invention. FIG. 2B is a flowchart showing a processing procedure in the first embodiment of the present invention. FIG. 3 is a diagram illustrating a chromosome that is an object of distance measurement and a nucleolus that is a non-object. FIG. 4 is a diagram showing one form of the cost field. FIG. 5 is a diagram showing one form of a dynamic route generation process. FIG. 6 is a block diagram showing the configuration of the distance measuring device in the second embodiment of the present invention. FIG. 7A is a flowchart showing a processing procedure in the second embodiment of the present invention. FIG. 7B is a flowchart showing a processing procedure in the second embodiment of the present invention. FIG. 8 is a diagram illustrating a line segment that is an extension of a route, a deleted line segment, and a non-object region.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Image input device 2 ... Image binarization device 3 ... Parameter setting device 4 ... Cost field generation device 5 ... Dynamic route generation device 6 ... Route search device 7 ... Route display device 8 ... Route reconstruction device 11 ... Chromosome ( Binarized image cluster)
12 ... Nucleolus (binarized image cluster)
13 ... Line segment to be a new extension of the route 14 ... Deleted line segment 15 ... Non-object area

Claims (11)

In a path measuring device that measures a path between the objects input from an image for an object for which a path candidate is not determined in advance,
An image input device that inputs an object of distance measurement and an image of a non-object excluded from the path of distance measurement;
An image binarization device that separates the inside and outside of the object and non-object regions included in the input image by binarization;
A parameter setting device that sets a start point that is a start point of the route, a destination point that is an end point of the route, a non-object region through which the route cannot pass, and an extension parameter r of the route;
Based on the start point, the target point, and the non-object region, the closer the start point is, the farther away from the target point, the greater the value, and the greater the value within the non-object region. A cost field generating device that defines a cost that defines the ease of passage of the route at each point of the input image, and generates a cost field that is a set of points where the cost is defined;
Generating a starting point set having the starting point as a single element, and generating a line segment set of line segments of length r extending from each starting point included in the starting point set; Update the existing start point set with a set of minute end points as a new start point set, and generate the start point set and the line segment set until all paths represented by the line segment set reach the destination point A dynamic route generation device that generates a set of the routes by sequentially repeating updates;
A route search device for searching for the shortest route from the routes included in the set of routes;
A route measuring device comprising: the shortest route; and a route display device that displays a length of the shortest route. In the route measuring device according to claim 1,
After obtaining the shortest path, it is checked whether the length r of the line segment satisfies a given condition. If the condition is not satisfied, the length of the line segment is reduced and the length of the shortest path is reduced. A path measurement device further comprising a path reconstruction device that holds a neighborhood as a non-object in the binary image. In the route measuring device according to claim 1 or 2,
In the process of generating the route, the dynamic route generation device is configured such that the distance between the end point of a line segment constituting one route and the end point of a line segment constituting another route becomes smaller than a given value. A route measuring device that integrates these end points into one when there is a failure. In the route measuring device according to any one of claims 1 to 3,
The route search device determines a route by a Viterbi algorithm. In the path | route measuring device as described in any one of Claims 1 thru | or 4,
The route search device determines a route by a Dijkstra algorithm. In the route measurement method that measures the route between objects for which no route candidate is determined in advance,
A first step of inputting an object of distance measurement and an image of a non-object excluded from the path of distance measurement;
A second step of separating the inside and the outside of the region of the object and the non-object included in the input image by binarization;
A third step of setting a start point that is a start point of the route, a destination point that is an end point of the route, a non-object region through which the route cannot pass, and an extension parameter r of the route;
Based on the start point, the target point, and the non-object region, the closer the start point is, the farther away from the target point, the greater the value, and the greater the value within the non-object region. A cost defining the ease of passing the route at each point of the input image, and generating a cost field that is a set of points where the cost is defined;
Generating a starting point set having the starting point as a single element, and generating a line segment set of line segments of length r extending from each starting point included in the starting point set; Update the existing start point set with a set of minute end points as a new start point set, and generate the start point set and the line segment set until all paths represented by the line segment set reach the destination point A fifth step of sequentially repeating the update to generate the set of the routes;
A sixth step of searching for the shortest route from the routes included in the set of routes;
A route measurement method comprising: the seventh step of displaying the shortest route and the length of the shortest route. The route measurement method according to claim 6,
After obtaining the shortest path, it is checked whether the length r of the line segment satisfies a given condition. If the condition is not satisfied, the length of the line segment is reduced and the length of the shortest path is reduced. And further comprising an eighth step of holding the vicinity as a non-object in the binary image,
A path measurement method that repeats the third, fourth, fifth, and sixth steps until the given condition is satisfied. In the route measuring method according to claim 6 or 7,
In the fifth step, in the process of generating the route, the distance between the end point of the line segment constituting one route and the end point of the line segment constituting the other route is smaller than a given value. A route measurement method that integrates these end points into a single case. In the route measurement method according to any one of claims 6 to 8,
A route measuring method for determining a route by a Viterbi algorithm in the sixth step. In the route measuring method according to any one of claims 6 to 9,
A route measurement method for determining a route by a Dijkstra algorithm in the sixth step.   The program for making a computer perform the path | route measurement method as described in any one of Claims 6 thru | or 10.

Images