WO2013172491A1

WO2013172491A1 - Method and device for recognizing object by using depth information

Info

Publication number: WO2013172491A1
Application number: PCT/KR2012/003840
Authority: WO
Inventors: 최병호; 김제우; 황영배; 배주한
Original assignee: 전자부품연구원
Priority date: 2012-05-16
Filing date: 2012-05-16
Publication date: 2013-11-21

Abstract

A method and device for recognizing an object by using depth information is provided. The method for recognizing the object includes: obtaining a color image and a depth image; clustering the color image in a plurality of clusters on the basis of depth information represented on the depth image; and performing object recognition on the basis of the clustered clusters. Accordingly, even if an object has few characteristic points in characteristic point-based object recognition or has a complex background, it is possible to more reliably increase an object recognition rate.

Description

【Specification】

[Name of invention]

Object recognition method and device using depth information

Technical Field

The present invention relates to an object recognition method and apparatus, and more particularly, to an object recognition method and apparatus using depth information.

Background Art

Feature point matching is mainly used as a method for recognizing whether an object to be found exists in a captured image.

This technique is able to find out the existence of the object to be found on the image as well as the location of the object relatively accurately, but it reveals the problem of inaccuracy in some cases.

Specifically, false object recognition may occur due to the background, which is more so, the more complicated the background. In addition, even when the feature descriptor has low discrimination power or when the recognition object is uniform, an incorrect feature point may appear, causing incorrect recognition. ^

On the other hand, since feature point matching for object recognition is performed for the entire image, it causes a reduction in recognition speed due to an increase in throughput, and there is a problem that the speed decrease will be further increased if the resolution of the image is increased.

[Detailed Description of the Invention] ^"

[Technical problem] The present invention has been made to solve the above problems, and an object of the present invention is to more reliably improve the object recognition rate even in the case of a feature point-based object recognition or an object having a complex background. It is to provide a method and apparatus for recognizing an object that can be increased.

Technical Solution

In accordance with an aspect of the present invention, an object recognition method includes: obtaining a color image and a depth image; Clustering the color-image into a plurality of clusters based on the depth information indicated in the depth-image; And performing object recognition based on the clustered clusters.

The object recognition may include: performing feature point matching on each of the clusters and each of the recognition objects; Selecting some regions in the clusters as candidate regions; And eliminating feature point matching made outside clusters including the selected candidate regions, and performing object recognition.

Further, performing object recognition may include selecting some regions in the clusters as candidate regions; Performing feature point matching on each of the candidate areas and each of the recognition objects; And performing object recognition using the feature point matching result.

The candidate region selecting step may include: a first selecting step of selecting some regions in the clusters as candidate regions by comparing color histograms for each of the clusters and color histograms for each of the recognition objects; And comparing the depth statistics of the candidate region selected in the first selection step with the recognition object, And a second selecting step of selecting a candidate region and a recognition object having a difference less than a threshold.

The object recognition step may further include: a first selecting step of selecting some areas in the clusters as candidate areas; Performing feature point matching on each of the candidate regions selected in the first selecting step and each of the recognition objects; A second selecting step of selecting again some of the candidate areas selected in the first selecting step; And eliminating feature point matching made outside the candidate regions selected in the second selection step, and performing object recognition.

The method may further include processing the recognized object recognized in the plurality of clusters as not recognized.

In addition, when a plurality of recognition objects are recognized in the color-image, comparing the color-correlation between the cluster and the recognition objects, selecting one recognition object and processing the recognized object.

On the other hand, the object recognition apparatus according to another embodiment of the present invention, the camera for acquiring a color-image and a depth-image; A storage unit in which images of recognition objects are DBized; And a processor that clusters the color-image into a plurality of clusters based on the depth information shown in the depth-image, and performs object recognition based on the clustered clusters.

Advantageous Effects

As described above, according to the present invention, even in the case where there are few feature points of an object in an object-based object recognition or an object having a complex background, it is more reliable It is possible to increase the object recognition rate.

Specifically, clustering an image using depth information, selecting candidate regions in a cluster through color-histogram comparison, and using the image recognition to increase the object recognition rate, as well as the amount of computation required for feature point matching and subsequent processing. It can be greatly reduced.

In addition, by comparing the depth statistics between the candidate area and the recognition object, the candidate area can be verified, thereby further increasing the object recognition rate.

In addition, since the similarity is further measured by using the color correlation as an auxiliary index for similar objects that may be difficult to recognize even using depth information, the recognition rate may be further increased.

[Brief Description of Drawings]

1 is a flowchart provided to explain an object recognition method using depth information according to an embodiment of the present invention;

2 shows color—image, depth-image and Depth Proximity Clustering results.

3 is a diagram showing three candidate regions selected from clusters in a color image.

4 is a diagram illustrating depth standard deviation of recognition objects;

5 and 6 are views illustrating a case where one recognition object is recognized in a plurality of clusters,

7 and 8 illustrate a case where a plurality of recognition objects are recognized in a color-image. Illustrated drawing ，

9 is a flowchart provided to explain an object recognition method using depth information according to an embodiment of the present invention;

10 is a flowchart provided to explain an object recognition method using depth information according to another embodiment of the present invention, and

FIG. 11 is a diagram illustrating an object recognition apparatus capable of performing the object recognition method illustrated in FIGS. 1, 9, and 10.

Best Mode for Implementation of the Invention

Hereinafter, with reference to the drawings will be described the present invention in more detail.

1. Object recognition using depth information # 1

The object recognition method using depth information according to an embodiment of the present invention uses depth information in clustering color-images and verifying a region selected as a candidate having a high probability of the object to be recognized in the cluster.

1 is a flowchart provided to explain an object recognition method using depth information according to an exemplary embodiment of the present invention.

As shown in FIG. 1, first, a color-image and a depth-image are obtained (S110). Color- and depth-images can be generated using RGB cameras and D-cameras, respectively.

Thereafter, the color-image is clustered into a plurality of clusters based on the depth information indicated in the depth-image (S120). Strictly speaking, clustering in the S120 phase In addition to the depth information, location information is further reflected, and the reflectance is greater in the depth information. Here, the position information refers to information on the position (x, y) on the plane perpendicular to the depth (d).

Clustering performed in step S120 is to cluster pixels having similar depths and close distances to one cluster. .

By the step S120, the 'object to be recognized' (hereinafter, abbreviated as 'aware object') existing on the color image is very likely to belong to any one cluster. This is because the pixels of one recognition object have similar depths and close distances. Recognition object is the object to find in the color-image, and the image is stored in DB.

Next, feature point matching is performed on each of the clusters and each of the recognition objects (S130). If k clusters are clustered and the number of objects recognized in the DB is _n , step S130 is performed.

11) perform feature point matching between the cluster -1 and the recognition object -1,

12) perform feature point matching between the cluster -1 and the recognition object -2,

In) perform feature point matching between the cluster -1 and the recognition object- _n ,

21) perform feature point matching between cluster -2 and recognition object-1,

22) perform feature point matching between the cluster -2 and the recognition object -2,

2n) perform feature point matching between cluster -2 and recognition object -n; kl) perform feature point matching between the cluster -k and the recognition object -1,

k2) perform feature point matching between the cluster ᅳ k and the recognition object-2;

kn) performed by a procedure for performing feature point matching between the cluster -k and the recognized object -n.

After that, the clustering clusters may first select the cluster clusters as candidate regions (S 140).

Step S140 is performed for each of the clustered clusters in step S120, and candidate region selection is performed by comparing the color and histogram of the cluster and the recognition object.

Specifically step S140,

11) If there is an area in the cluster # 1 that has a color-histogram similar to the color-histogram of the recognition object-1, the area is selected as a candidate area,

12) In the cluster -1, if there is an area with a color-histogram similar to the color-histogram of the object-2, the area is selected as a candidate area.

In) If there is an area with a color-histogram similar to the color-histogram of the recognition object -n in cluster -1, the area is selected as a candidate area,

21) In the cluster -2, if there is a region with a color-histogram similar to the color-histogram of the recognition object-1, the region is selected as a candidate region,

22) In the cluster -2, if there is a region with a color-histogram similar to the color histogram of the object # 2, the region is selected as a candidate region, 2n) If there is an area with a -histogram similar to the color-histogram of recognition object -n in cluster -2, the area is selected as a candidate area, and kl) with the color-histogram of recognition object -1 in cluster -k. If there is an area with a similar color-histogram, the area is selected as a candidate area, and k2) if there is an area with a color- histogram similar to the color- histogram of the recognition object -2 in the cluster -k, the area. Is selected as a candidate region, and if there is a region having a color- histogram similar to the color_ histogram of the recognition object -n in kn) cluster -k, the region is selected as the candidate region. Next, the depth standard deviations between the candidate region firstly selected in step S140 and the recognition object are compared, and the candidate regions are secondarily selected (S150).

Here, the depth standard deviation is the standard deviation of the depth values for the pixels, the depth standard deviation of the candidate area is the standard deviation of the depth values of the pixels constituting the candidate area, and the depth standard deviation of the recognition object is an image of the recognition object. The standard deviation of the depth values of the pixels constituting

The depth standard deviation of the recognition object is preferably calculated in advance and stored in the DB. On the other hand, it is also possible to replace the standard deviation with other statistics, for example, variance, mean, median, and the like.

In step S140, 1) Some areas of cluster -1 were selected as 'candidate area -1' because 'recognition object -1' and color-histogram are similar.

2) A partial region of the cluster -3 was selected as the candidate region -2 because the recognition object -Γ and the color histogram are similar.

3) Some areas of cluster -2 were selected as 'candidate area -3' because 'cognitive object -2' and color- histogram are similar.

4) Some areas of the cluster -4 were selected as the candidate area -4 because the recognition object -2 and the color histogram are similar.

5) Some areas of cluster -3 were selected as 'candidate area -5' because the recognition object -3 and the color ᅳ histogram are similar.

6) Some areas of cluster -5 were selected as 'candidate area -6' because 'cognitive object -4' and color- histogram are similar.

7) Assuming that some areas of the cluster -6 are selected as the candidate area -7 because the recognition object -4 and the color-histogram are similar.

In S 150

1) If the difference between the depth standard deviation of the candidate area -1 and the depth standard deviation of the recognition object # 1 is less than the threshold value, the candidate area-1 and the recognition object # 1 are selected.

2) If the difference between the depth standard deviation of the candidate area -2 and the depth standard deviation of the recognition object -1 is less than the threshold value, the candidate area -2 and the recognition object # 1 are selected.

3) If the difference between the depth standard deviation of the candidate area -3 and the depth standard deviation of the recognition object -2 is less than the threshold value, the candidate area -3 and the recognition object -2 are selected. 4) If the difference between the depth standard deviation of the candidate area -4 and the depth standard deviation of the recognition object -2 is less than the threshold value, the candidate area -4 and the recognition object -2 are selected.

5) If the difference between the depth standard deviation of the candidate area -5 and the depth standard deviation of the recognition object -3 is less than the threshold value, the candidate area -5 and the recognition object-3 are selected.

6) If the difference between the depth standard deviation of the candidate area -6 and the depth standard deviation of the recognition object -4 is less than the threshold, select the candidate area -6 and the recognition object-4,

7) If the difference between the depth standard deviation of the candidate area -7 and the depth standard deviation of the recognition object -4 is less than the threshold value, the candidate area # 7 and the recognition object -4 are selected.

Subsequently, the object recognition is performed by excluding feature point matching for clusters that do not include candidate regions selected in step S150 from the feature point matching result in step S130 (S160). That is, the object recognition is performed while leaving only the feature point matching for the clusters including the candidate regions selected in step S150 among the feature point matching results in step S130.

In step S150, 'candidate area -1 and recognition object -1', 'candidate area -2 and recognition object -1', 'candidate area -3 and recognition object -2', 'candidate area -5 and recognition object -3' 'And' candidate area— 6 and recognition object -4 'are selected,

Candidate region -1 is included in cluster -1, candidate region -3 is included in cluster -2, candidate region -2 and candidate region -5 are included in cluster -3, and candidate region -6 is included in cluster -5 Included,

In step S160, only feature point matching for cluster-1, cluster-2, cluster-3, and cluster # 5 is left, except for feature point matching for the remaining clusters, Perform object recognition

On the other hand, when one recognized object is recognized in a plurality of clusters, the recognized object may be treated as not recognized (S170). Specifically, a recognition object in which the ratio of the feature point matching number between the first cluster with the highest feature point matching number and the second largest cluster within a specific range is treated as not recognized.

For example, the second matching number is the same as in the case where the recognition object—1 has a feature point matching number (①) of cluster −1 is “100” and a feature point matching number (②) of cluster −3 is “90”. If (②) is 80% or more of the first matching number (①) and the recognition object -1 can be treated as recognized in both cluster -1 and cluster -3, the recognition object -1 is treated as not recognized. .

The same recognition object is recognized in multiple clusters because it is more likely to be due to an error in the image recognition process than it actually is, and to prevent false recognition in advance.

In addition, when a plurality of recognition objects are recognized in at least one cluster of the color-image, only one recognition object is selected by comparing the color correlation between the two clusters with the highest feature point matching number and the recognition objects, and the selected recognition is performed. The object is processed as recognized (S180).

For example, the feature point matching number ③ of the recognition object -2 and the cluster -2 is "90", the feature point matching number (④) of the recognition object -3 and the cluster -3 is "60", and the recognition object -4 The second matching number (⑤) is more than 70% of the first matching number (③), as in the case where the characteristic point matching number (⑤) of the cluster-5 is " 70 " Object -4 All of them can be treated as recognized.

In this case, the color-correlation of recognition object -2 and cluster-2 is compared with the color-correlation of recognition object -4 and cluster -5. It's what's recognized in the image.

If the feature point matching number ③ of the recognition object—2 and the cluster-2 is “90”, the feature point matching number ④ of the recognition object −3 and the cluster −3 is “40”, and the recognition object −4 and the cluster If the second matching number (④) is less than 70% of the first matching number (③), such as the case where the characteristic point matching number (⑤) of -5 is "30", the matching number is the most without the color correlation. Many recognition objects -2 treat as recognized in cluster -2 of color-images.

So far, the entire process of the object recognition method using depth information has been described in detail. Hereinafter, each step of configuring the object recognition method will be described in detail.

2. Depth Proximity Clustering (S120 in FIG. 1)

Depth Proximity Clustering refers to clustering in step S120. In order to reduce the inaccuracy of feature point matching caused by a complex background, color-image is divided into an object and a background and clustered.

Clustering may be performed using K-means clustering all, and similarly close and close distances may be clustered in one cluster with reference to depth information and location information. Specifically, the clustering may be performed according to Equation 1 below.

[Equation 1] k

arg _s min ^ H ^x i-^ II

i Xj ^ Sj

Where k is the number of clustering, is the geometric center of ¾ of the i th cluster, and Xj is a vector expressed by [cx, a ₂ y, a ₃ d]. x and y are the x and y coordinates of the pixel, and d is the depth value. a _1; a ₂ and a ₃ are weights, and are preferably set to (^ = 0.05, a ₂ = 0.05 and a ₃ = 0.9 to reflect the effect of depth more than the position of the pixel (x ᅳ y). Of course, it can be changed accordingly.

In the upper left part of FIG. 2, a color-image of a specific space is shown (in case of inevitably converted to a hog-image due to the limitation of drawing standards), and in the lower left, a depth-image of the specific space is shown. Shows the Depth Proximity Clustering results.

3.Color Histogram based Object Detection (S140 of FIG. 1)

When an object is recognized in a complex background with a uniform object that does not have many feature points, a lot of faulty feature point matching appears. Therefore, in the present embodiment, the object detection algorithm using color histogram back projection is used to select a region where a recognition object is likely to exist even in a cluster as a candidate region.

In FIG. 3, three candidate regions selected by Color Histogram based Object Detection for the clusters shown in the clustering result of FIG. Indicated).

4.Equality Test of Standard Deviation (SI 50 of FIG. 1)

The Equality Test of Standard Deviation can be understood as a procedure for verifying or reselecting candidate areas where recognition objects are likely to exist by comparing depth standard deviations between candidate areas selected through Color Histogram based Object Detection and recognition objects. have.

Figure 4 shows the depth standard deviation for the recognition objects, books, dolls, cushions. As shown in FIG. 4, the depth standard deviation for an object mostly follows a Gaussian distribution regardless of the shape of the object. Therefore, assuming that the random sample of the recognition object and the candidate area follows a normal distribution, the ratio of the standard deviation follows the F distribution. And, the reliability is α, the degrees of freedom are η _χ -1, n _y -l, the degrees of freedom have the following confidence interval when set to the number of random samples, matching beyond the confidence interval is removed and the rest is reliable It is considered a match.

Confidence interval: (F _{α / 2} , η _χ -1, _1 F _Q / 2, n _x -l, n _y -l)

5.Depth Layered Feature Matching with Geometric Constraints (FIG. 1

S160)

As a feature point matching in step S130, it is preferable to use a feature point matching technique that is invariant to scale and rotation, and to use another type of feature point matching technique. It is not excluded.

On the other hand, when not considering the radial distortion caused by the lens, in general, the relationship between the two images can be expressed as a matrix. In the image, when there are several large points on the same plane as the building wall, it can be approximated to exist on one plane and this relationship is called Homography.

The homography between the cluster and the recognition object generated through Depth Proximity Clustering is more accurate than the homography between the color-image and the recognition object because the matching of the feature point of the recognition object and the background is excluded.

Subsequently, homography between clusters containing candidate regions selected through 1 or ⁰ Color Histogram based Object Detection and Equality Test of Standard Deviation (S140 and S150 in FIG. Because it is more excluded, it is more accurate than homography between clusters and recognition objects.

6. Final decision with consistency check (S170 and SI 80 of £ 1)

Final decision with consistency check is a procedure for final object recognition when one recognition object is recognized in multiple clusters and when multiple recognition objects are recognized in a color-image.

The case where one recognition object is recognized in a plurality of clusters is illustrated in FIGS. 5 and 6. As shown in FIGS. 5 and 6, it can be seen that the recognition object "cushion" shows a large number of feature point matchings in different clusters. As mentioned above As such, if the ratio of the number of feature point matches between both is within a certain range, the "cushion" in the color-image is treated as not recognized. 7 and 8 illustrate a case where a large number of recognition objects are recognized in a color image (represented by inevitably converting to a hog-image due to limitations in drawing standards). Feature point matching for "book" and "doll" with the top two matching numbers is shown. If the ratio of feature point matching between the two is outside the specified range, for example, the feature number matching between "book" and cluster is equal to "doll". If the number of feature point matches between clusters is significantly greater than the number of feature points matching between clusters, the "book" is considered recognized, whereas if the ratio of the number of feature points matching between the two is within a certain range, the color_correlation between cluster and "book" and the cluster and "doll" By comparing the color-correlation between the two objects, a recognition object with a large color-correlation is recognized as being recognized in a cluster 7. Object recognition using depth information # 2

Hereinafter, a description of an object recognition method using depth information according to another exemplary embodiment of the present invention will be described with reference to FIG. 9, and the same parts as the above description will be briefly described. As shown in FIG. 9, first, a color-image and a depth-image are obtained (S210). Thereafter, the color-image is clustered into a plurality of clusters based on the depth information indicated in the depth-image (S220).

Thereafter, some regions in the clusters are first selected as candidate regions (S230). In operation S230, each of the clustered clusters is performed in operation S220, and the candidate region selection is performed by comparing the color and histogram of the cluster and the recognition object.

Next, feature point matching is performed on each of the candidate areas and each of the recognition objects (S240). The feature point matching in step S240 of FIG. 9 is a feature point matching between the "candidate area" and the recognition object, which is different from the feature point matching between the "cluster" and the recognition object in step S130 of FIG.

Subsequently, candidate regions similar in depth standard deviation between the candidate region firstly selected in step S230 and the object to be recognized are secondarily selected (S250).

Next, the feature point matching is excluded from the feature point matching in step S240 and the object point matching performed outside the "candidate regions" selected in step S250 is performed (S260). That is, object recognition is performed while leaving only the feature point matchings included in the candidate areas selected in step S250 among the feature point matching results in step S240.

Feature point matching excluded in step S260 is different from step S160 in FIG. 1 to exclude feature point matching made outside clusters including the second selected candidate areas in that the feature point matchings are made outside the second selected candidate areas. .

On the other hand, when one recognized object is recognized in a plurality of "candidate areas", the recognized object may be treated as not recognized (S270). Specifically, a recognition object in which the ratio of the feature point matching number between the first candidate region with the greatest number of feature point matches and the second candidate region is within a specific range is treated as not recognized.

In addition, in the at least one candidate region of the color-image, a plurality of recognition objects If it is recognized, only one recognition object is selected by comparing the color-correlation between the candidate regions having the top two feature point matching numbers and the recognition objects, and the selected recognition object is processed as being recognized (S280).

Specifically, in step S280, only the candidate region having the ratio of the feature point matching number between the first recognition object with the highest number of feature points and the second recognition object within a specific range is performed. Otherwise, the recognition object with the highest number of feature point matches is treated as recognized.

The processing criteria in steps S270 and S280 are different from those in steps S170 and S180 of FIG. 1, in which the processing criteria are clusters.

8. Object Recognition Using Depth Information # 3

Hereinafter, a description will be given of an object recognition method using depth information according to another embodiment of the present invention with reference to FIG. 10.

In the object recognition method illustrated in FIG. 10, after performing primary and secondary selection of candidate regions (S330 and S340), feature point matching is performed on each of the secondary selected candidate regions and each of the recognition objects (S330 and S340). In operation S350, the object is recognized in operation S360, which is different from the object recognition method illustrated in FIG. 9, and the rest thereof are the same, and thus a detailed description thereof will be omitted.

9.Object recognition device

FIG. 11 may perform the object recognition method illustrated in FIGS. 1, 9, and 10. It is a figure which shows the object recognition apparatus. As shown in FIG. 11, the object recognition apparatus to which the present invention is applicable includes an RGB camera 210, a D-camera 420, a processor 430, a storage 440, and a display 450. .

RGB-camera 410 produces a color-image, and D-camera 420 produces a depth-image. The RGB-camera 410 and the D-camera 420 may be configured separately as shown in FIG. 11, but may also be configured as one camera.

In the storage unit 440, images of recognition objects are DBized. On the other hand, the storage unit 440 is also a DB of the depth standard deviations for the recognition objects. Processor 430 is a color-image generated by the RGB camera 410,

The depth-image generated by the D-camera 420 and the storage unit 440 perform the object recognition method illustrated in FIGS. 1, 9, and 10 by using the stored recognition objects.

Display 450 shows a color-image generated by RGB camera 410, which may be displayed as a stereo-image using depth-image generated by D-camera 420. In addition, the display 450 displays an object recognition result performed by the processor 430. In addition, although the preferred embodiment of the present invention has been shown and described above, the present invention is not limited to the specific embodiments described above, the technical field to which the invention belongs without departing from the spirit of the invention claimed in the claims Of course, various modifications can be made by those skilled in the art, and these modifications can be individually understood from the technical spirit or the prospect of the present invention. I will not.

Claims

[Range of request]

[Claim 1]

Obtaining a color-image and a depth-image;

Clustering the color-image into a plurality of clusters based on the depth information indicated in the depth-image; And

Based on the clustered clusters, performing object recognition.

[Claim 2]

The method of claim 1,

The object recognition step is

Performing feature point matching on each of the clusters and each of the recognition objects; Selecting some regions in the clusters as candidate regions; And

And performing object recognition excluding feature point matching made outside clusters including the selected candidate regions.

[Claim 3]

The method according to claim 1,

The object recognition step is

Selecting some regions in the clusters as candidate regions; Performing feature point matching on each of the candidate areas and each of the recognition objects;

And performing object recognition using the feature point matching result.

[Claim 4]

The method according to claim 2 or 3,

Candidate area selection step is

A first selecting step of comparing the color histograms for each of the clusters with the color histograms for each of the recognition objects, and selecting some regions in the clusters as candidate regions; And

And a second selection step of comparing the depth statistics of the recognition region and the recognition object selected in the first selection step, and selecting the candidate region and the recognition object having a statistical difference less than a threshold value.

[Claim 5]

The method of claim 1,

The object recognition step is

A first selecting step of selecting some regions in the clusters as candidate regions;

Performing feature point matching on each of the candidate regions selected in the first selecting step and each of the recognition objects; A second selecting step of selecting again some of the candidate areas selected in the first selecting step; And

And excluding object point matching made outside of the candidate regions selected in the second selection step, and performing object recognition.

[Claim 6]

The method of claim 1,

And processing the recognized object recognized in the plurality of clusters as not recognized.

[Claim 7]

The method of claim 1,

If a plurality of recognition objects are recognized in the color-image, comparing the color-correlation between the cluster and the recognition objects, selecting one recognition object, and processing the recognized object as a recognized object; Recognition method.

[Claim 8]

A camera for acquiring color- and depth-images;

A storage unit in which images of recognition objects are made into a DB; And

Depth-Based on the depth information shown in the image, Cluster into clusters, based on clustered clusters

And a processor for performing the object recognition apparatus.