JP2019036030A - Object detection device, object detection method and object detection program - Google Patents

Abstract

To improve detection accuracy of a cylindrical body or a spherical body form an image.SOLUTION: An object detection device 1 detects a cylindrical or spherical object from an image, and includes a projection transformation unit 11 which performs projection transformation of an original image which is obtained by imaging an object from an inclined angle with respect to a plane in a space into a front image obtained by imaging the object from a direction perpendicular to the plane, a foreground region extraction unit 12 which extracts a foreground region located in front of the plane from the front image, a designated color extraction unit 13 which extracts a designated color region matching the designated color from the foreground region, a profile detection unit 14 which approximates a profile surrounding the designated color region with an ellipse, and a shape verification unit 15 which determines whether the ellipse is a circle or not, and outputs the ellipse as a cylindrical object of a spherical object if the ellipse is a circle.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a technique for detecting a cylindrical or spherical object from an image.

  Currently, in sports competitions and surveillance fields, the position of an object that exists in real space is identified from video captured by a camera, and further, the position of the object is tracked. Specifically, research is being conducted on a technique for detecting a cylindrical or spherical object existing in space from at least one image input from a camera that captures real space, and outputting information on the object to the screen. Has been developed.

  For example, Non-Patent Document 1 discloses a technique for detecting a cylindrical object arranged using the Mean shift method of Non-Patent Document 2 so that the normal direction of the bottom surface is parallel to the normal line of the floor plane. ing. In the mean shift method, feature values are extracted from a specified region that is a template, a region having a feature value similar to this is detected from the image currently being processed, and the position corresponding to the template is detected from the image being processed. Estimate as In many cases, a histogram created from color information in a specified area is used as the feature amount. However, erroneous detection may occur when a similar color pattern exists in the background or the like. Therefore, in Non-Patent Document 1, a highly accurate detection of a cylindrical object is realized by using a histogram created by adding edge information to color information.

J. Kim, 3 others, “Curling Stone Tracking by an Algorithm Using Appearance and Color Features”, Proceedings of the World Congress on Electrical Engineering and Computer Systems and Science (EECSS 2015), Paper No. 334, July 2015 D. Comaniciu, 2 others, “Real-time Tracking of Non-rigid Objects Using Mean Shift”, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2015, p.142-p.149

  In the conventional method, a cylindrical object arranged so that the normal direction of the bottom surface is parallel to the normal line of the floor plane is detected using a histogram created from edge information in the designated area.

  However, when a shape pattern similar to the cylindrical object to be detected exists in the background, there is a problem that the detection target is erroneously detected by using only the edge information. In addition, since geometric information is lost in the histogram representation, the histogram generated from an object having a shape pattern that is not a cylindrical shape that exists in the surrounding area is similar to that created from a cylindrical object, so that the detection target There is also a problem that false detection occurs.

  Another problem is that if the orientation of the camera that captured the current image is different from the orientation of the camera when the template image was captured, the orientation and strength of the edges extracted from the cylindrical object are Since they are different, there is a problem in that detection of a cylindrical object is leaked. When a cylindrical object is photographed so that the direction of the cylinder axis coincides with the optical axis of the camera, it is observed as a perfect circle on the photographed image. The size is much different from the perfect circle. When the former is used as a template and the latter is used as an input image, erroneous detection of a cylindrical object occurs.

  The present invention has been made in view of the above circumstances, and when there is a shape pattern similar to the shape of a cylinder or sphere as a subject in the background or surroundings, or when the normal direction of the cylinder bottom and the optical axis of the camera are Even in the case of an input image taken from a direction that does not match, a cylinder in which the normal direction of the bottom surface existing in the space is known and parallel to the bottom surface from at least one image obtained by photographing the subject The object is to accurately detect a body or a sphere and output information on the object.

  In order to solve the above problems, an object detection apparatus according to claim 1 is an object detection apparatus that detects a cylindrical or spherical object from an image, and is an original image taken from an oblique angle with respect to a plane in space. Is specified from the foreground area, a projection conversion section that performs projective conversion to a front image photographed from an angle perpendicular to the plane, a foreground extraction section that extracts a foreground area located in front of the plane from the front image, and A designated color extraction unit that extracts a designated color region that matches a color; a contour detection unit that approximates an outline surrounding the designated color region by an ellipse; and determines whether the ellipse is a circle, and the ellipse is a circle In the case of (3), a shape verification unit that outputs the ellipse as a cylindrical or spherical object is provided.

  The object detection device according to claim 2 is the object detection device according to claim 1, wherein the position of the ellipse is inversely converted to the position of the coordinate system before projective transformation, and the plane of the cylinder or sphere forming the ellipse. A detection position correction unit is further provided that adds a predetermined offset value based on the height from the position and then returns to the position of the coordinate system after the projective transformation.

  The object detection method according to claim 3 is an object detection method for detecting a cylindrical or spherical object from an image, wherein a computer captures an original image photographed at an oblique angle with respect to a plane in space. Projective conversion to a front image taken from a perpendicular angle, extracting a foreground region located in front of the plane from the front image, and extracting a specified color region that matches a specified color from the foreground region A step of approximating an outline surrounding the specified color area with an ellipse, and determining whether the ellipse is a circle. If the ellipse is a circle, the ellipse is output as a cylinder or a spherical object. And a step of performing.

  According to a fourth aspect of the invention, an object detection program causes a computer to function as the object detection device according to the first or second aspect.

  According to the present invention, it is possible to accurately detect a cylinder or a sphere from an image.

It is a figure which shows the example of the real space and input image which are used by each embodiment. It is a figure which shows the functional block structure of an object detection apparatus. It is a figure which shows the flowchart of an object detection process. It is a figure which shows the flowchart of a projective transformation process. It is a figure which shows the example of an input image. It is a figure which shows the example of a background image. It is a figure which shows the example of a background geometric transformation image. It is a figure which shows the example of an input geometric transformation image. It is a figure which shows the flowchart of a foreground extraction process. It is a figure which shows the example of a foreground area | region image. It is a figure which shows the flowchart of a designated color extraction process. It is a figure which shows the example of a designated color extraction image. It is a figure which shows the flowchart of an outline detection process. It is a figure which shows the example of an outline extraction image. It is a figure which shows the example of an ellipse approximate image. It is a figure which shows the flowchart of a shape verification process. It is a figure which shows the example of the shift | offset | difference of the center position between coordinate systems by the height of a cylindrical object. It is a figure which shows the functional block structure (2nd Embodiment) of an object detection apparatus. It is a figure which shows the flowchart (2nd Embodiment) of a projective transformation process. It is a figure which shows the flowchart of a detection position correction process. It is a figure which shows the example of an offset value. It is a figure which shows the example of a conversion of the position coordinate from a projective transformation image coordinate system to an input image coordinate system. It is a figure which shows the example of a conversion of the position coordinate from an input image coordinate system to a projective transformation image coordinate system.

  As described above, an object of the present invention is to improve the detection accuracy of a cylinder or a sphere from an image. For this reason, the present invention projects and transforms an original image taken from an oblique angle with respect to a plane in space into a front image taken from an angle perpendicular to the plane, Is extracted from the foreground area, the specified color area that matches the specified color is extracted from the foreground area, the outline surrounding the specified color area is approximated by an ellipse, and it is determined whether the ellipse is circular. When the ellipse is circular, the ellipse is output as a cylindrical or spherical object.

  That is, in the present invention, since the original image taken from an oblique angle with respect to the plane in the space is projectively converted to the front image, the shape of the subject can be grasped in its original shape, and the detection target cylinder The detection omission of a body or a sphere can be suppressed, and detection can be performed reliably. In the present invention, the foreground region is extracted from the front image, the object is detected from the foreground region using the color and the contour information, and it is verified whether or not the shape of the detected object is a circle. It is possible to suppress overdetection in which an object other than a body or a sphere is detected together. As a result, it is possible to improve the detection accuracy of the cylinder or sphere from the image.

  Hereinafter, an object detection device, an object detection method, and an object detection program according to an embodiment will be described with reference to the drawings. In each embodiment to be described later, in the real space of the (x ′, y ′, z ′) coordinate system having the x′-y ′ plane as the floor as shown in FIG. From the input image taken by the camera 3 in which the z ′ axis and the optical axis A do not match, the normal direction of the bottom surface of the cylinder in real space is the positive direction of the z ′ axis perpendicular to the floor surface, and the cylinder The cylindrical objects 5a, 5b, and 5c existing on the floor surface that is a plane parallel to the bottom surface are detected as target objects.

  For example, it can be used when detecting stones on ice used in curling competitions. In addition to the cylindrical object, a sphere used in a billiard game or a gateball game may be used as a target object.

  In addition, the said input image can be paraphrased in other words as the original image image | photographed from the diagonal angle with respect to x'-y 'plane in real space.

<First Embodiment>
FIG. 2 is a diagram illustrating a functional block configuration of the object detection device 1 according to the first embodiment. The object detection device 1 is a device that detects a cylindrical or spherical object from an image, and as shown in FIG. 2, a projective transformation unit 11, a foreground extraction unit 12, a designated color extraction unit 13, and a contour. A detection unit 14, a shape verification unit 15, and a memory 16 are provided.

  The projective transformation unit 11 receives at least one or more input images, calculates a projective transformation matrix, geometrically transforms a predetermined background image using the calculated projective transformation matrix, and stores it in the memory 16 as a background geometric transformed image. And a functional unit that records and outputs an input geometric conversion image by geometrically converting the input image using the projective transformation matrix.

  The foreground extraction unit 12 inputs the input geometric transformation image and the background geometric transformation image, and is positioned in front of the x′-y ′ plane by background difference processing using the input geometric transformation image and the background geometric transformation image. This is a functional unit that extracts a foreground area and outputs the extracted foreground area as a foreground area image.

  The designated color extraction unit 13 inputs a foreground area image, inputs predetermined color information and a color range, and extracts a specified color area similar to the predetermined color within the color range from the foreground area image. It is a functional unit that outputs as an extracted image.

  The contour detection unit 14 inputs a designated color extraction image, extracts a contour surrounding each designated color region over all the designated color regions included in the designated color extraction image, and approximates the extracted contour with an ellipse. It is a functional part that outputs as ellipse information.

  The shape verification unit 15 inputs all the ellipse information, inputs a predetermined roundness threshold value, and performs roundness threshold processing for each ellipse information to determine whether the ellipse being processed is close to a circle. When the ellipse being processed is close to a circle, the ellipse information of the ellipse is output as object information.

  Next, object detection processing performed by the object detection apparatus 1 will be described using the flowchart of FIG.

Step S101;
First, the projective transformation unit 11 is activated. Projection conversion processing performed by the projection conversion unit 11 will be described with reference to the flowchart of FIG.

Step S101a;
At least one image is input to the projective conversion unit 11. In the present embodiment, for example, an input image shown in FIG. 5 is input. The projection conversion unit 11 inputs predetermined four corresponding points when the activation is the first time. In the present embodiment, as the corresponding points, for example, (x ₁ , y ₁ ) → (x ′ ₁ , y ′ ₁ ), (x ₂ , y ₂ ) → (x ′ ₂ , y ′ ₂ ), (x ₃ , Y ₃ ) → (x ′ ₃ , y ′ ₃ ), (x ₄ , y ₄ ) → (x ′ ₄ , y ′ ₄ ) are designated. The symbol “→” indicates the correspondence between the real space and the input image. (X ′ ₁ , y ′ ₁ ), (x ′ ₂ , y ′ ₂ ), (x ′ ₃ , y ′ ₃ ), (x ′ ₄ , y ′ ₄ ) are respectively (x ₁ , y ₁ ) , (X ₂ , y ₂ ), (x ₃ , y ₃ ), (x ₄ , y ₄ ), points in the real space, and points on a plane parallel to the bottom surface of the cylinder.

Step S101b;
Next, the projective transformation unit 11 calculates a projective transformation matrix H. The projective transformation matrix is also called homography, and is a matrix that represents a geometric transformation from a reference image to a subject in the input image when the subject is a plane, and is represented by, for example, a 3 × 3 matrix. The projective transformation matrix can be obtained if there are four or more corresponding points (identical points) between the recognition target and the reference image in the input image. The projective transformation unit 11 calculates a projective transformation matrix H using the above-described four corresponding points.

Steps S101c and S101d;
Thereafter, the projective transformation unit 11 performs geometric transformation of a predetermined background image using the calculated projective transformation matrix H, and records it in the memory 16 as a background geometric transformation image. In the present embodiment, it is assumed that the background image shown in FIG. 6 is given in advance, and that the background geometric transformation image shown in FIG.

Steps S101e, S101f;
Finally, the projective transformation unit 11 performs geometric transformation on the input image using the projective transformation matrix H as in step S101c, records the input geometric transformation image in the memory 16 and finishes the process. In the present embodiment, it is assumed that the input geometric transformation image shown in FIG.

  As a result, an input image obtained by photographing a cylindrical object from a direction in which the normal direction of the bottom surface of the cylinder and the optical axis of the camera do not match is projected onto an image captured from a direction in which the normal direction of the bottom surface of the cylinder matches the optical axis of the camera. Will be converted. In other words, an original image captured from an oblique angle with respect to the x′-y ′ plane in the space is projectively transformed into a front image captured from an angle perpendicular to the plane.

Step S102 of FIG. 3;
Next, the foreground extraction unit 12 is activated. The foreground extraction process performed by the foreground extraction unit 12 will be described with reference to the flowchart of FIG.

Step S102a;
The foreground extraction unit 12 inputs the background geometric transformation image and the input geometric transformation image read from the memory 16. In the present embodiment, the background geometric transformation image shown in FIG. 7 and the input geometric transformation image shown in FIG. 8 are input.

Step S102b;
Next, the foreground extraction unit 12 performs background difference processing using the input two converted images. The background difference process can be realized by obtaining an absolute value of a difference between pixels corresponding to the same position in the background geometric transformation image and the input geometric transformation image.

Step S102c;
Next, the foreground extraction unit 12 performs smoothing processing on the difference image because the difference image obtained by the background difference processing often includes fine noise. The smoothing process can be realized, for example, by applying a Gaussian filter to the difference image. When the noise included in the difference image is small, the smoothing process may be skipped giving priority to the processing speed.

Step S102d;
Next, the foreground extraction unit 12 obtains a mask image indicating the position of the foreground pixels by performing binarization processing with a predetermined threshold on the difference image.

Step S102e;
Next, the foreground extraction unit 12 extracts foreground pixels from the input geometric transformation image and the mask image. This process can be realized, for example, by an AND operation process of the input geometric transformation image and the mask image.

Step S102f;
Finally, the foreground extraction unit 12 records an image composed of the extracted foreground pixels as a foreground area image in the memory 16 and ends the process. In the present embodiment, it is assumed that the foreground area image shown in FIG. 10 is obtained. As shown in FIG. 10, a background pattern that may cause a false detection is removed in advance in step S102.

Step S103 of FIG. 3;
Next, the designated color extraction unit 13 is activated. The designated color extraction process performed by the designated color extraction unit 13 will be described with reference to the flowchart of FIG.

Step S103a;
The designated color extraction unit 13 inputs predetermined color information and a foreground area image. The predetermined color information is a value indicating the color of an object to be detected, and has two values, a minimum value and a maximum value, for each predetermined color. For example, when the object to be detected is blue, the lowest value (0,0,200) and the highest value (0,0,255) are displayed in RGB. In the case of red, the lowest value (200, 0, 0) and the highest value (255, 0, 0) are similarly expressed in RGB. For simplicity of explanation, the description has been made using RGB notation, but it may be expressed in other color spaces such as HSV, HLS, and LAB. In that case, the color space of the foreground area image is also converted.

Step S103b;
Next, the designated color extraction unit 13 extracts a region corresponding to the designated color from the foreground region image and the color information. This process can be realized by performing extraction for all pixels for a pixel in a foreground area image when the pixel value falls between the maximum value and the minimum value of the color information. An image obtained as a result of being performed on all pixels is referred to as a designated color extraction image.

Step S103c;
Next, the specified color extraction unit 13 may change the color information on the image due to the influence of illumination in the real environment, and the extracted pixels are often missing or excessively extracted. Smoothing processing is performed on the specified color extraction image. The smoothing process can be realized, for example, by applying a Gaussian filter or a median filter, or applying a morphology operation. Similarly to step S102c, the smoothing process may be skipped giving priority to the processing speed.

Step S103d;
Finally, the designated color extraction unit 13 records the designated color extraction image in the memory 16 and finishes the process. In the present embodiment, it is assumed that there are two colors of blue and red as the colors of the object to be detected, and the above-described RGB values are used as the color information. As a result, the images shown in FIGS. 12A and 12B are obtained as the designated color extraction images.

Step S104 of FIG. 3;
Next, the contour detection unit 14 is activated. The contour detection process performed by the contour detection unit 14 will be described with reference to the flowchart of FIG.

Step S104a;
The contour detection unit 14 inputs a designated color extraction image.

Step S104b;
Next, the contour detection unit 14 converts the input designated color extraction image into a gray scale, and binarizes it with a predetermined threshold value.

Step S104c;
Next, the contour detection unit 14 extracts a contour that surrounds each of the designated color regions included in the binarized image. This processing can be realized, for example, by the method of a reference (Suzuki, S, 1 other, “Topological Structural Analysis of Digitized Binary Images by Border Following”, CVGIP 301, 1985, p.32-p.46). . In this embodiment, the contours shown in FIGS. 14A and 14B are obtained for each of the two designated color extraction images.

Step S104d;
Finally, the contour detection unit 14 approximates all the extracted contours with ellipses, outputs the approximate ellipse ellipse information, and ends the processing. Here, the ellipse information is information including at least the center coordinates of the ellipse, the major axis and minor axis of the ellipse, and the rotation angle of the ellipse. The process of approximating an outline with an ellipse is described in, for example, a reference document (Andrew W, 1 other, “A Buyer's Guide to Conic Fitting”, Proc. 5th British Machine Vision Conference, Birmingham, 1995, p.513-p.522. ). In the present embodiment, ellipses shown in FIGS. 15A and 15B are obtained for each of the two designated color extraction images.

Step S105 in FIG. 3;
Next, the shape verification unit 15 is activated. The shape verification process performed by the shape verification unit 15 will be described with reference to the flowchart of FIG.

Step S105a;
The shape verification unit 15 inputs all the ellipse information and a predetermined roundness threshold value.

Steps S105b to S105d;
Next, the shape verification unit 15 determines whether or not the ellipse being processed is close to a circle by performing roundness threshold processing on each of the ellipse information. The roundness threshold processing can be measured, for example, based on whether the ratio of the minor axis to the major axis of the ellipse is close to 1.0. In this case, the roundness threshold value processing is determined to be close to a circle if the minor axis / major axis> threshold value. All ellipses determined to be close to a circle are output as object information, and the process ends. In this embodiment, 5a, 5b, and 5c shown in FIGS. 15A and 15B output the center coordinates, the major and minor diameters of the ellipse, and the rotation angle of the ellipse as circular object information. Since 5d that approximates a human shape by an ellipse is not nearly circular, the object information is not output.

  The first embodiment has been described above.

  According to this embodiment, an input image obtained by photographing a cylindrical object from a direction in which the normal direction of the bottom surface of the cylinder and the optical axis of the camera do not match is captured from a direction in which the normal direction of the bottom surface of the cylinder matches the optical axis of the camera. Since projective conversion into an image is performed, the shape of the subject can be grasped in its original shape, and the detection omission of the cylinder or sphere that is the detection target can be suppressed and reliably detected.

  Further, according to the present embodiment, a foreground area is extracted from a projective-transformed image, an object is detected from the foreground area using color and contour information, and whether or not the shape of the detected object is circular is verified. Therefore, it is possible to suppress over-detection in which objects other than cylindrical objects are detected together.

  As a result, it is possible to improve the detection accuracy of the cylindrical object from the input image. Even in the case of a sphere, the shape after projective transformation is circular as in the case of a cylindrical object. Therefore, even if the input image includes a sphere, the object of the sphere can be detected with high accuracy. .

<Second Embodiment>
In the first embodiment, even in the case of an input image taken from a direction in which the normal direction of the bottom surface of the cylinder and the optical axis of the camera do not match, the normal direction of the bottom surface of the cylinder and the optical axis of the camera are converted by projective transformation processing. Projective transformation from the matching direction to an image taken, and detecting an object from the projected image using color and contour information, and verifying the shape of the detected result, the subject is a cylindrical object In addition, it realized that the detection was performed accurately.

However, the projective transformation process used in the first embodiment is based on the premise that the subject is a flat body having no height. For example, when the real space shown in FIG. 5 is targeted, it is surrounded by (x ₁ , y ₁ ), (x ₂ , y ₂ ), (x ₃ , y ₃ ), and (x ₄ , y ₄ ). Although an accurate projective transformation process can be performed on the plane portion, a cylindrical object or a person existing on the plane is distorted in the image after the projective transformation.

  Therefore, if the upper surface of the cylindrical object is higher than the floor plane composed of the four points specified for projective transformation matrix calculation, the detection position and the position on the plane will be misaligned. become.

  On the other hand, for example, when it is desired to measure the exact position of the stone in a curling competition, it may be necessary to specify the position of the cylindrical object on the plane that becomes the game field. Therefore, in the present embodiment, as shown in FIG. 17, the deviation of the detection position due to the height of the cylindrical object or sphere is eliminated, and the center position of the detected cylindrical object is used to perform projection conversion processing on the target plane. Estimate the center position.

  FIG. 18 is a diagram illustrating a configuration block configuration of the object detection device 1 according to the second embodiment. As shown in FIG. 18, the object detection device 1 is configured by adding a detection position correction unit 17 to the configuration in the first embodiment.

  The detection position correction unit 17 is a functional unit that receives the object information output from the shape verification unit 15, corrects the position of the detected object, and outputs the object information after position correction.

  Next, the object detection process performed by the object detection apparatus 1 will be described. In the present embodiment, each process of the projection conversion unit 11 that performs processing different from that of the first embodiment and the detection position correction unit 17 added in the second configuration mode will be described.

  First, the projective transformation process performed by the projective transformation unit 11 will be described using the flowchart of FIG.

  At least one image is input to the projective conversion unit 11. In the present embodiment, for example, the input image shown in FIG. 5 is input. In this embodiment, steps S205 and S206 are added to the processing of the first embodiment (FIG. 4). Hereinafter, description of the same steps (steps S201 to S204, S207 to S208) as in the first embodiment will be omitted, and only steps S205 and S206 will be described.

Step S205;
The projective transformation unit 11 calculates an inverse matrix H ⁻¹ from the projective transformation matrix H calculated in step S202. The inverse matrix H ⁻¹ is a matrix expressing a geometric transformation from a projective transformed image to an input image. For example, H ⁻¹ × (x ′ ₁ , y ′ ₁ , 1) ^ T = (x ₁ , y ₁ , 1).

Step S206;
Next, the projective transformation unit 11 records the calculated projective transformation matrix H and the inverse matrix H ⁻¹ of the projective transformation matrix H in the memory 16.

  Next, processing performed by the detection position correction unit 17 will be described using the flowchart of FIG.

Step S301;
The detection position correction unit 17 inputs all object information that has undergone the shape verification process and a predetermined offset. Here, as shown in FIG. 21, the offset refers to the center position on the plane targeted by the projective transformation process from the center position of the object on the input image in consideration of the height of the cylindrical object, for example. (Predetermined offset value based on the height of the cylindrical object from the x′-y ′ plane) and can be expressed as a vector d passing through both points, for example.

Step S302;
Next, the detection position correction unit 17 performs the subsequent processing for a certain object.

Step S303;
First, the detection position correcting unit 17 inversely transforms the coordinates of the center position of the object into the coordinates of the input image coordinate system before projective transformation using the inverse matrix H- ¹ . In the present embodiment, as shown in FIG. 22, the coordinates of the center position of the detected object are (c _x ′, _cy ′) and are converted into (c _x , _cy ) by the inverse matrix H ^−1. Shall be.

Step S304;
Next, the detection position correction unit 17 adds a predetermined offset to the converted coordinates. In the present embodiment, it is assumed that (d _x , _dy ) is given in advance as the offset d, and the coordinates after the addition are (c _x + d _x , _cy + d _y ).

Step S305;
Next, the detection position correction unit 17 reconverts the coordinates (c _x + d _x , c _y + d _y ) into the projective transformation image coordinate system using the projection transformation matrix H. In the present embodiment, as shown in FIG. 23, ((c _x + d _x ) ′, (c _y + d _y ′ ′)) is calculated as the estimation result of the center position on the target plane in the projective transformation process. become.

Step S306;
Next, the detection position correction unit 17 updates the re-transformed coordinate value as a new coordinate of the center position of the object to be processed, and stores it in the memory 16.

Step S307;
The detection position correction unit 17 performs the above processing on all the input objects, and finally outputs all the object information stored in the memory 16 to finish the processing.

  The second embodiment has been described above.

  According to the present embodiment, the coordinates of the center position of the object are inversely converted to the coordinates of the input image coordinate system, and from the center position of the object on the input image to the center position on the target plane in the projective transformation process. After adding the offset value, the coordinates after adding the offset value are returned to the coordinates of the projective transformation image coordinate system, so that the upper surface of the cylindrical object is higher than the plane composed of the four points designated for the projective transformation matrix calculation. Even if it exists in the place, the center position on the plane can be estimated, and for example, in a curling competition, the exact position of the stone on the plane that becomes the game field can be calculated.

  As described above, the present invention has been specifically described based on the example of the embodiment. However, the above description of the embodiment is for explaining the present invention, and limits the invention described in the claims. Or it should not be construed to reduce the scope. Moreover, each means structure of this invention is not restricted to the above-mentioned embodiment, Of course, various deformation | transformation are possible within the technical scope as described in a claim. For example, the foreground extraction process in step S102 shown in FIG. 3 may be performed before the projective conversion process in step S101.

  Finally, the object detection apparatus 1 described in each embodiment can be realized by a computer or the like having the above functions. It is also possible to create an object detection program for causing a computer to function as the object detection apparatus 1 and a storage medium for the object detection program.

DESCRIPTION OF SYMBOLS 1 ... Object detection apparatus 11 ... Projection conversion part 12 ... Foreground extraction part 13 ... Designated color extraction part 14 ... Contour detection part 15 ... Shape verification part 16 ... Memory 17 ... Detection position correction part

Claims (4)

In an object detection device for detecting a cylindrical or spherical object from an image,
A projective transformation unit that projectively transforms an original image taken from an oblique angle with respect to a plane in space into a front image taken from an angle perpendicular to the plane;
A foreground extraction unit that extracts a foreground region located in front of the plane from the front image;
A designated color extraction unit that extracts a designated color area that matches the designated color from the foreground area;
An outline detector that approximates an outline surrounding the specified color area with an ellipse;
It is determined whether or not the ellipse is a circle, and if the ellipse is a circle, a shape verification unit that outputs the ellipse as a cylinder or a sphere object;
An object detection apparatus comprising:   The position of the ellipse is inversely converted to the position of the coordinate system before projective transformation, and after adding a predetermined offset value based on the height of the cylinder or sphere forming the ellipse from the plane, the coordinate system after projective transformation The object detection apparatus according to claim 1, further comprising a detection position correction unit that returns the position to the position. In an object detection method for detecting a cylindrical or spherical object from an image,
Computer
Projectively transforming an original image taken from an angle oblique to a plane in space into a front image taken from an angle perpendicular to the plane;
Extracting a foreground region located in front of the plane from the front image;
Extracting a designated color area that matches a designated color from the foreground area;
Approximating an outline surrounding the specified color area with an ellipse;
Determining whether the ellipse is circular, and if the ellipse is circular, outputting the ellipse as a cylindrical or spherical object;
The object detection method characterized by performing.   An object detection program for causing a computer to function as the object detection apparatus according to claim 1.

Images