KR20150104857A - Object identification apparatus, method thereof and computer readable medium having computer program recorded therefor - Google Patents

Abstract

The present invention discloses an apparatus for identifying an object, a method thereof, and a recording medium having a computer program recorded therefor. In other words, when learning a key point for identifying an object, the apparatus calculates feature information for an image by applying a feature pattern wherein a region is divided in the shape of a circle or a ring for the image and performs learning and object identification based on the calculated feature information, thereby improving a feature extracting performance for the object including a curved surface.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an object identification apparatus, an object identification apparatus, a method thereof, and a recording medium on which a computer program is recorded.

The present invention relates to an object identification apparatus, a method thereof, and a recording medium on which a computer program is recorded. More particularly, the present invention relates to a method and apparatus for calculating feature information on an image by applying a feature pattern, An object identification device for performing learning and object identification on the basis of the feature information, a method thereof, and a recording medium on which a computer program is recorded.

Learning method Object identification is based on features (or feature points) obtained from an image (or image), and expresses each feature with various types of descriptors.

This learning method object identification uses a Harr-like feature method. Since the corresponding halftone feature method is based on a quadrangle, in order to search an object (or object) such as a person or a vehicle It requires a lot of training.

Korean Patent No. 10-0752251 [Title: Object selection method and system]

An object of the present invention is to provide an object identifying device for calculating feature information on an image by applying a feature pattern in which regions are divided in the form of a circle or a ring for an image, a method thereof, and a recording medium on which a computer program is recorded .

Another object of the present invention is to provide an object identification device for calculating feature information on an image through a randomly preset (or arranged) feature map for a feature pattern divided into zones in the shape of a circle or a ring, And a recording medium on which the recording medium is recorded.

The object identification apparatus according to the embodiment of the present invention applies a circular or donut-shaped characteristic pattern divided into a plurality of zones to an image, and based on the preset characteristic map corresponding to the characteristic pattern, A control unit for selecting a zone corresponding to a preset reference among the zones and calculating position information of a zone selected for each of the plurality of unit items included in the feature map as image feature information of an area where the feature pattern is located in the image; And a storage unit for storing image feature information of an area where the calculated feature pattern is located.

As an example related to the present invention, a preset reference may be the brightest or darkest zone among the zones corresponding to the unit item.

As an example related to the present invention, the controller may repeat the process of selecting a zone corresponding to a reference for each of a plurality of unit items included in the feature map.

As an example related to the present invention, when calculating image feature information of a region in which a feature pattern is located, zone information of a feature pattern corresponding to a zone selected for each of the plurality of unit items is divided into image feature information .

As an example related to the present invention, the controller may apply a circular or donut-shaped feature pattern to all the remaining regions of the image while sliding, and calculate a plurality of region-specific image feature information for a plurality of regions of the entire image, Learning can be performed on the basis of a plurality of region-by-region image feature information for the entire image.

An object identification apparatus according to an exemplary embodiment of the present invention includes a storage unit for storing a plurality of image feature information for each of a plurality of images; And a degree of similarity between a plurality of image feature information for each area calculated by applying a feature pattern of a circular or donut shape to the image and a plurality of image feature information for each region for a plurality of images stored in the storage unit, And a controller for determining whether the object exists in the image based on the similarity.

As an example related to the present invention, the controller may determine whether the determined similarity degree is equal to or greater than a preset reference value.

As an example related to the present invention, when the identified similarity degree is equal to or greater than a preset reference value, the controller can determine whether or not a target object exists in the image and identify the object in the image.

As an example related to the present invention, when the determined similarity degree is smaller than a preset reference value, it can be determined that there is no object in the image.

According to an embodiment of the present invention, there is provided a method for identifying an object, comprising: applying a feature pattern of a circle or a donut shape divided into a plurality of regions to an image through a control unit; Selecting a zone corresponding to a preset reference among the zones included in the unit item of the feature map based on a preset feature map corresponding to the feature pattern through the control unit; Calculating position information of a zone selected for each of the plurality of unit items included in the feature map through the control unit from the image as image feature information of the region where the feature pattern is located; And storing image feature information of an area in which the feature pattern calculated in the storage unit is located.

As an example related to the present invention, there is provided a method for searching a feature map, the method comprising: repeatedly performing a process of selecting a region corresponding to a reference for each of a plurality of unit items included in a feature map through a controller; Calculating a plurality of image feature information for each of a plurality of regions of the entire image by applying a circular or toroidal feature pattern to all remaining regions of the image through a control unit; And performing a learning based on a plurality of image feature information for each region of the entire image calculated through the control unit.

The step of calculating image feature information of an area where the feature pattern is located in the image may include calculating a feature corresponding to a zone selected for each of the plurality of unit items, The zone information of the pattern can be calculated as the image feature information of the region where the feature pattern is located.

According to an embodiment of the present invention, there is provided a method for identifying an object, the method comprising: calculating a plurality of image feature information for each region by applying a circular or toroidal feature pattern to the image through a controller; Checking the similarity between a plurality of image feature information for each region calculated through the control unit and a plurality of image feature information for each region for a plurality of images stored in the storage unit; And determining whether a target object exists in the image based on the similarity determined through the control unit.

According to an embodiment of the present invention, the step of determining whether an object exists may include determining whether the degree of similarity determined through the control unit is equal to or greater than a preset reference value; Determining whether a target object exists in the image through the control unit when the determined similarity is equal to or greater than a preset reference value, and identifying the object in the image; And determining that the target object does not exist in the image through the control unit when the determined similarity is smaller than a preset reference value.

A computer program for carrying out the method according to the above-described embodiment may be stored in the recording medium on which the computer program according to the embodiment of the present invention is recorded.

The present invention has an effect of improving feature extraction performance for an object including a curved surface by calculating feature information on an image by applying a feature pattern in which regions are divided in a circle or ring shape with respect to the image.

Further, according to the present invention, feature information on an image is calculated through a feature map that is randomly preset (or arranged) with respect to a feature pattern divided into regions in the shape of a circle or a ring, And the area where the feature can be extracted at a time is wide, so that there is an effect of performing a fast process.

1 is a block diagram illustrating a configuration of an object identification apparatus according to an embodiment of the present invention.
Figures 2 to 4 show FIG. 8 is a diagram illustrating an example of a feature pattern according to an embodiment of the present invention.
5 is a diagram illustrating an example of a feature map according to an embodiment of the present invention.
6 is a diagram illustrating an example of a feature map and feature information according to an embodiment of the present invention.
Figure 7 1 is a flowchart illustrating an object identification method according to a first embodiment of the present invention.
8 is a diagram illustrating an image according to an embodiment of the present invention.
9 is a cross- 5 is a flowchart illustrating an object identification method according to a second embodiment of the present invention.

It is noted that the technical terms used in the present invention are used only to describe specific embodiments and are not intended to limit the present invention. In addition, the technical terms used in the present invention should be construed in a sense generally understood by a person having ordinary skill in the art to which the present invention belongs, unless otherwise defined in the present invention, Should not be construed to mean, or be interpreted in an excessively reduced sense. In addition, when a technical term used in the present invention is an erroneous technical term that does not accurately express the concept of the present invention, it should be understood that technical terms that can be understood by a person skilled in the art can be properly understood. In addition, the general terms used in the present invention should be interpreted according to a predefined or prior context, and should not be construed as being excessively reduced.

Furthermore, the singular expressions used in the present invention include plural expressions unless the context clearly dictates otherwise. The term "comprising" or "comprising" or the like in the present invention should not be construed as necessarily including the various elements or steps described in the invention, Or may further include additional components or steps.

Furthermore, terms including ordinals such as first, second, etc. used in the present invention can be used to describe elements, but the elements should not be limited by terms. Terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, wherein like reference numerals refer to like or similar elements throughout the several views, and redundant description thereof will be omitted.

In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. It is to be noted that the accompanying drawings are only for the purpose of facilitating understanding of the present invention, and should not be construed as limiting the scope of the present invention with reference to the accompanying drawings.

1 is a block diagram showing the configuration of an object identifying apparatus 10 according to an embodiment of the present invention.

As shown in FIG. 1, the object identification apparatus 10 includes a storage unit 100 and a control unit 200. Not all of the components of the object identification apparatus 10 shown in Fig. 1 are essential components, and the object identification apparatus 10 may be implemented by more components than the components shown in Fig. 1, The object identifying device 10 may also be implemented by a component.

The control unit 200 applies a circular or toroidal (or ring-like) feature pattern divided into a plurality of regions to an image (or image). Thereafter, the controller 200 selects a region corresponding to a preset reference among the regions corresponding to the unit items included in the feature map that is set in advance in correspondence with the feature pattern. Then, the controller 200 calculates the position information of the plurality of zones, which are selected for each of the plurality of unit items, as the image feature information of the region where the feature pattern is located in the image. In addition, the control unit 200 slides and applies a circular or donut-shaped (or ring-like) feature pattern divided into a plurality of regions to all the remaining regions of the image, A plurality of image feature information for each area). Thereafter, the control unit 200 learns based on the plurality of image feature information for each region of the original image. In addition, the control unit 200 performs an object identification function for a plurality of image feature information for each region calculated from other images based on the plurality of image characteristic information for each region of the plurality of learned images.

The storage unit 100 stores a user interface (UI), a graphical user interface (GUI), and the like.

In addition, the storage unit 100 stores data and programs necessary for the object identification apparatus 10 to operate.

The storage unit 100 stores an image (or an image) captured through an input unit (not shown) (or a camera (not shown)) included in the object identification apparatus 10, (Not shown), and the like.

The storage unit 100 stores a feature pattern of a circular or toroidal shape (or ring shape) divided into a plurality of zones, a preset feature map corresponding to the feature pattern, a plurality of (or more) Item, one or more zone information selected from a plurality of zones included in a characteristic pattern of a circle or a donut included in the unit item, and the like. Here, the circular or toroidal feature patterns can be set to various sizes according to the designer's design so that the feature information can be calculated for a plurality of original images of different sizes.

Also, the storage unit 100 stores a plurality of image feature information, a learning result, an identified (or recognized) object information for an image generated by the control unit 200, and the like.

The control unit 200 executes the overall control function of the object identification device 10. [

The controller 200 also applies a circular or donut-shaped (or ring-shaped) feature pattern divided into a plurality of regions to an image (or image) stored in the storage unit 100 or an image received through the communication unit or the input unit .

That is, as shown in FIG. 2, the controller 200 applies a circular feature pattern 210 to an image. At this time, the feature pattern is not limited to the circle shown in FIG. 2, but may be a figure including a curve according to the designer's design, such as the donut shape 310 shown in FIG. 3, the double donut shape 410 shown in FIG. It is possible to design and change variously.

Also, the control unit 200 sets a feature map corresponding to the feature pattern. Here, the feature map may be mutually shared among different object identification devices in order to extract feature information on the image. In addition, the feature map used in the object identifying apparatus 10 may be a feature map provided (shared) from another object identifying apparatus.

Also, as shown in FIG. 5, the feature map 500 includes a plurality of (or more) unit items 510. Here, each unit item 510 is divided into a plurality of zones 511, 512, and 513, and each zone 511, 512, and 513 includes a zone selected from a plurality of zones included in a circular or donut- (Or settings) 521, 522, and 523. At this time, the number of unit items included in the feature map and the number of zones included in the unit item can be variously set according to the designer's design.

In addition, the controller 200 selects a zone corresponding to a preset reference (or appropriate / satisfactory) among the zones corresponding to the unit items included in the feature map, based on a preset feature map corresponding to the feature pattern . Here, the preset reference includes information for confirming whether the region is the brightest region or the darkest region among the regions included in the unit item. At this time, when there are a plurality of unit items, the control unit 200 repeatedly performs a process of selecting a zone corresponding to a preset reference for each of a plurality of unit items.

6, the controller 200 selects the regions 611, 612, 613, 614, and 615 corresponding to the brightest region (or the darkest region) do.

In addition, the control unit 200 calculates (or sets) the position information of each of the plurality of (or more than one) unit items as the image feature information of the region where the feature pattern is located.

That is, as shown in FIG. 6, the controller 200 determines the position information 620 of each of the five unit items (for example, the area / area hatched by each unit item) As shown in Fig.

In the embodiment of the present invention, it is described that the image feature information of the region in which the feature pattern is located is a combination of position information in which each selected region of the unit item is located. However, the present invention is not limited thereto. The image feature information of the region may be a combination of region information of the feature pattern corresponding to each selected region of the unit items.

That is, five regions of the feature pattern corresponding to the brightest region (or the darkest region) among the regions of (0, 2, 5) included in the first unit item shown in FIG. 6 are selected, A region of the feature pattern corresponding to the brightest region among the regions of (1, 4, 7) included is selected, and a region corresponding to the brightest region among the regions of (2, 3, 6) 6 zones of the feature pattern corresponding to the brightest region among the zones of (4, 6, 7) included in the fourth unit item are selected and the 6 zones of the feature pattern corresponding to the The controller 200 selects one of the five unit patterns, that is, the zone information of the selected feature pattern (for example, five zones, one zone , 3 zone, 6 zone, 5 zone), the corresponding circle The features in the pattern may be calculated by the image feature information of the partial region.

In addition, the control unit 200 slides and applies a feature pattern of a corresponding circle or donut shape to all the remaining regions of the image, and applies the feature pattern to the plurality of regions of the entire original image, ) (Or extracts / sets).

In addition, the control unit 200 stores a plurality of image feature information for each region of the original image in the storage unit 100.

In addition, the controller 200 learns based on a plurality of image characteristic information for each area of the calculated image. At this time, in the case of the corresponding image, the control unit 200 may normalize the image to a plurality of predetermined sizes, and then calculate a plurality of image-specific image-specific information for each of the plurality of normalized images.

In addition, the controller 200 may perform learning based on a plurality of image feature information for each of the plurality of images.

In addition, the control unit 200 stores the learning result in the storage unit 100. [

The control unit 200 may further include a circular or donut-shaped (e.g., rectangular) shape divided into a plurality of zones for another image (or image) stored in the storage unit 100 or another image received through a communication unit (not shown) A plurality of image feature information for each region is calculated by applying the feature pattern of the region. In addition, the control unit 200 stores a plurality of image feature information for each region in the storage unit 100.

In addition, the control unit 200 may store each of a plurality of image feature information for each region for a plurality of images stored (or learned / registered) in advance in the storage unit 100 and a plurality of images for each region (Or compares) the degree of similarity between feature information.

In addition, the control unit 200 determines whether the similarity between the plurality of image feature information of each region calculated and the plurality of images stored in the storage unit 100 is equal to or greater than a preset reference value (Or confirms).

That is, the control unit 200 determines whether each of the plurality of similarity degrees detected is equal to or greater than a preset reference value.

As a result of the determination, if there is a similarity that is equal to or greater than a preset reference value among a plurality of similarity levels that have been confirmed, classification (or determination / verification) is made as a target object exists in the corresponding image corresponding to the similarity degree, (Or recognizes) the identified object, and stores the identified object in the storage unit 100.

If it is determined that there is no similarity that is equal to or greater than a preset reference value among the plurality of similarity levels that have been checked, that is, if all of the plurality of similarity levels are smaller than a preset reference value, the control unit 200 determines that no object exists in the corresponding image Classify.

In this manner, feature information about an image can be calculated by applying a feature pattern in which regions are divided into a circle or a ring shape with respect to the image.

In this way, feature information on an image can be calculated through a feature map that is randomly preset (or arranged) with respect to a feature pattern in which regions are divided in the shape of a circle or a ring.

Hereinafter, an object identification method according to the present invention will be described in detail with reference to FIGS. 1 to 9. FIG.

Figure 7 1 is a flowchart illustrating an object identification method according to a first embodiment of the present invention.

First, the control unit 200 applies a circular or donut-shaped (or ring-like) feature pattern divided into a plurality of zones to a partial area of the original image. Here, the feature pattern has a predetermined size.

For example, as shown in FIG. 8, the control unit 200 applies a circular feature pattern 820 divided into a plurality of regions to a partial region 810 of the original image (S710).

Thereafter, the controller 200 selects a region corresponding to (or suitable to) a predetermined criterion among the regions corresponding to the unit items included in the feature map, based on a preset feature map corresponding to the feature pattern. At this time, the controller 200 repeatedly performs a process of selecting a zone corresponding to a preset reference for a plurality of (or more than one) unit items included in the feature map. Here, the preset reference may be to check whether it is the brightest region or the darkest region among the regions included in the unit item.

5, the control unit 200 determines whether or not the characteristic map 500 included in the feature map 500 has been generated based on the preset feature map 500 corresponding to the circular feature pattern 210 shown in FIG. A zone corresponding to a preset reference among the zones 511, 512, and 513 corresponding to the unit item 510 is selected.

That is, as shown in FIG. 5, the feature map 500 is a feature map 500 in which the first unit item includes a region of (0, 2, 5) among the feature patterns of the circular shape, 3, 6) among the feature patterns of the circular shape and the fourth unit item includes the region of the feature patterns of the circle (4, 6, 7), and the fifth unit item is composed of a total of five unit items including a zone of (0, 3, 5) among the circular feature patterns. Also, the control unit 200 selects a region corresponding to the brightest region among the corresponding regions, which is a preset reference among the regions included in the five unit items. 6, the control unit 200 selects the second zone 611 having the brightest region 5 among the zones of 0, 2, 5 included in the first unit item (for example, 6), and the 0th region 612 having 1, which is the brightest region among the (1, 4, 7) regions included in the second unit item, is selected and included in the third unit item The first region 613 having the brightest region 3 is selected from among the regions of (4, 6, 7) included in the fourth unit item, and the brightest region 6 And selects the second zone 615 having the brightest region 5 among the zones (0, 3, 5) included in the fifth unit item (S720).

Then, the control unit 200 calculates (or sets) the positional information of each of the plurality of (or more than one) unit items as the image feature information of the region where the feature pattern is located in the corresponding image. At this time, the controller 200 may calculate the zone information of the feature pattern corresponding to the zone selected from the unit items as the image feature information of the region where the feature pattern is located in the corresponding image.

For example, as shown in FIG. 6, the control unit 200 determines the position of each of the five unit items (for example, the second zone, the zeroth zone, the first zone, the first zone, and the second zone) Information '20112' 620 is calculated as image feature information of a partial region in which a circular feature pattern is located (S730).

Then, the control unit 200 slides and applies the characteristic pattern of the circle or the donut type to all the remaining regions of the original image, and applies the characteristic pattern of the original image to the plurality of regions of the entire original image, Information) is extracted (or extracted / set).

In addition, the controller 200 stores a plurality of image feature information for each of the regions of the original image in the storage unit 100 (S740).

Thereafter, the control unit 200 learns based on the plurality of image feature information for each region of the original image.

In addition, the control unit 200 stores the learning result in the storage unit 100 (S750).

9 is a cross- 5 is a flowchart illustrating an object identification method according to a second embodiment of the present invention.

First, the control unit 200 displays a circle or a donut shape (or a circle or a circle) divided into a plurality of regions with respect to an image (or an image) stored in the storage unit 100 or an image received through a communication unit (not shown) or an input unit Ring type) feature pattern is applied to calculate a plurality of image feature information for each region.

In addition, the controller 200 stores a plurality of image feature information for each of the images in the storage unit 100 (S910).

Thereafter, the controller 200 calculates the similarity between the calculated image feature information of each of the plurality of regions and the plurality of image characteristic information of each region of the plurality of images stored (or learned / registered) in the storage unit 100 Confirm (or compare) each.

For example, the control unit 200 checks the similarity between the calculated image feature information of each of the plurality of regions and the image feature information of each of the plurality of regions with respect to the ◯◯ images previously learned in the storage unit 100 (S920 ).

Thereafter, the control unit 200 determines whether the similarity between the plurality of image feature information of each area calculated and the plurality of image characteristic information of each of the plurality of images stored in the storage unit 100 is equal to or greater than a preset reference value (Or confirms).

That is, the control unit 200 determines whether each of the plurality of similarity degrees detected is equal to or greater than a preset reference value.

For example, the control unit 200 determines whether the determined degree of similarity is equal to or greater than a preset reference value (S930).

As a result of the determination, if the determined similarity is equal to or greater than a preset reference value, the control unit 200 classifies (or judges / confirms) the object as a target object in the image, identifies (or recognizes) And stores the identified object in the storage unit 100.

For example, when there is a degree of similarity that is equal to or greater than a preset reference value out of the identified similarity degrees, the controller 200 determines a plurality of region-based image feature information , Identifies the object in the input image, and stores the identified object in the storage unit 100 (S940).

If the determined similarity is smaller than a preset reference value, the control unit 200 classifies the object as having no object in the image.

For example, when all of the determined similarity levels are less than a preset reference value, the control unit 200 classifies the object as not existing in the image (S950).

The object identification apparatus and method according to the embodiment of the present invention can be written in a computer program, and the codes and code segments constituting the computer program can be easily deduced by a computer programmer in the field. In addition, the computer program is stored in a computer readable medium readable by a computer and read and executed by a computer, an object identification device, a user device (or a terminal) according to an embodiment of the present invention, Device and method thereof.

The information storage medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium. The object identification apparatus according to an embodiment of the present invention and a computer program implementing the method can be stored and installed in an internal memory such as an object identification apparatus, a user apparatus, and the like. Alternatively, an external memory such as a smart card storing and installing an object identifying device according to an embodiment of the present invention and a computer program implementing the method may be mounted on an object identifying device, a user device, or the like through an interface.

As described above, in the embodiment of the present invention, feature information on an image is calculated by applying a feature pattern in which regions are divided in a circle or a ring shape with respect to the image, Can be improved.

In the embodiment of the present invention, as described above, feature information on an image is calculated through a feature map that is randomly preset (or arranged) with respect to a feature pattern in which regions are divided in the shape of a circle or a ring, It is possible to improve the identification efficiency of an object including a curved surface and to perform fast processing because the area where features can be extracted at a time is wide.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or essential characteristics thereof. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

The present invention is characterized in that, when learning a feature point for object identification, characteristic information about an image is calculated by applying a feature pattern in which regions are divided in a circle or a ring shape to an image, and learning and object identification are performed based on the calculated feature information The present invention can improve the feature extraction performance of an object including a curved surface and enhance the identification efficiency of an object including a curved surface and perform fast processing by broadening the area where features can be extracted at a time. Object identification / recognition field, and the like.

10: Object identifying apparatus 100:
200:

Claims (15)

A method of applying a feature pattern of a circle or donut shape divided into a plurality of zones to an image and setting a feature map corresponding to a preset reference among the zones included in the unit item of the feature map on the basis of a preset feature map corresponding to the feature pattern A controller for selecting a zone and calculating position information of a zone selected for each of the plurality of unit items included in the feature map from the image as image feature information of an area in which the feature pattern is located; And
And a storage unit for storing image feature information of an area where the calculated feature pattern is located. The method according to claim 1,
Wherein the predetermined criterion is the brightest or darkest zone among the zones corresponding to the unit item. The method according to claim 1,
Wherein the controller repeatedly performs a process of selecting a zone corresponding to the reference for each of the plurality of unit items included in the feature map. The method according to claim 1,
Wherein the control unit calculates the region information of the feature pattern corresponding to the region selected for each of the plurality of unit items as the image feature information of the region where the feature pattern is located when calculating the image feature information of the region in which the feature pattern is located The object identification apparatus comprising: The method according to claim 1,
Wherein the controller slides and applies the circular or toroidal feature pattern to all the remaining areas of the image to calculate a plurality of image feature information for each area of the entire image, Wherein the learning unit performs learning based on a plurality of region-by-region image feature information. A storage unit for storing a plurality of region-by-region image feature information for a plurality of images; And
The degree of similarity between a plurality of image feature information for each region calculated by applying a feature pattern of a circular or donut type to an image and a plurality of image feature information for each region for a plurality of images stored in the storage unit, And determining whether there is an object in the image based on the degree of similarity. The method according to claim 6,
Wherein the controller determines whether the determined similarity is equal to or greater than a preset reference value. 8. The method of claim 7,
Wherein the control unit determines whether a target object exists in the image when the determined similarity degree is equal to or greater than the predetermined reference value and identifies an object in the image. 8. The method of claim 7,
Wherein the controller determines that a target object does not exist in the image when the determined similarity is smaller than the preset reference value. Applying a circular or toroidal feature pattern divided into a plurality of zones to an image through a control unit;
Selecting a region corresponding to a preset reference among the regions included in the unit item of the feature map based on a preset feature map corresponding to the feature pattern through the control unit;
Calculating position information of a zone selected for each of a plurality of unit items included in the feature map through the controller as image feature information of an area where the feature pattern is located in the image; And
And storing image feature information of an area where the calculated feature pattern is located in the storage unit. 11. The method of claim 10,
Repeating the step of selecting a zone corresponding to the reference for each of the plurality of unit items included in the feature map through the control unit;
Calculating a plurality of image feature information for each of a plurality of regions of the entire image by applying the circular or toroidal feature pattern to all remaining regions of the image while sliding the feature pattern through the controller; And
And performing learning based on a plurality of image feature information for each region of the entire image through the control unit. 11. The method of claim 10,
Wherein the step of calculating image feature information of an area in which the feature pattern is located comprises the steps of calculating image feature information of an area where the feature pattern is located, Wherein zone information is calculated as image feature information of an area where the feature pattern is located. Calculating a plurality of image feature information for each region by applying a circular or toroidal feature pattern to the image through the control unit;
Confirming the similarity between the calculated image feature information for each region and the plurality of image feature information for each region for a plurality of images stored in the storage unit through the control unit; And
And determining whether a target object exists in the image based on the determined similarity through the control unit. 14. The method of claim 13,
Wherein the step of determining whether the object is present comprises:
Determining whether the determined similarity is equal to or greater than a preset reference value through the controller;
Determining whether a target object exists in the image through the control unit when the determined similarity is equal to or greater than the preset reference value and identifying an object in the image; And
And determining that a target object does not exist in the image through the control unit when the determined similarity is smaller than the preset reference value. 15. A recording medium on which a computer program for performing the method according to any one of claims 10 to 14 is recorded.

Images