KR20100101812A - Chain code representation method and apparatus for object motion analysis - Google Patents

Chain code representation method and apparatus for object motion analysis Download PDF

Info

Publication number
KR20100101812A
KR20100101812A KR1020090020204A KR20090020204A KR20100101812A KR 20100101812 A KR20100101812 A KR 20100101812A KR 1020090020204 A KR1020090020204 A KR 1020090020204A KR 20090020204 A KR20090020204 A KR 20090020204A KR 20100101812 A KR20100101812 A KR 20100101812A
Authority
KR
South Korea
Prior art keywords
direction vector
chain code
chain
position coordinates
image
Prior art date
Application number
KR1020090020204A
Other languages
Korean (ko)
Inventor
오춘석
유영기
한창호
Original Assignee
선문대학교 산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 선문대학교 산학협력단 filed Critical 선문대학교 산학협력단
Priority to KR1020090020204A priority Critical patent/KR20100101812A/en
Publication of KR20100101812A publication Critical patent/KR20100101812A/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G06T7/55—Depth or shape recovery from multiple images
    • G06T7/593—Depth or shape recovery from multiple images from stereo images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30241—Trajectory

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)

Abstract

PURPOSE: A chain code presentation method and an apparatus for the moving object analysis for expressing the movement of the object existing within a multi-dimensional image in a chain code are provided to improve the operation processing speed of the system for the moving object analysis by expressing the movement of the object. CONSTITUTION: An image implementation unit(100) implements a photographic image from a camera. The image implementation unit applies the photographic image from the camera. A code expression unit(200) detects the constellation of the object signal-changed according to the motion trajectory of the object existing within the multi-dimensional image.

Description

Chain Code Representation Method and Apparatus for Object Motion Analysis for Object Motion Analysis

The present invention relates to a chain code representation apparatus and method for object motion analysis, and more particularly, to a chain code representation apparatus and method for object motion analysis for expressing a movement of an object existing in a multi-dimensional image with a chain code.

Extracting and recognizing objects in an image or analyzing object movement using binary images uses less memory than using color or gray scale images that require a large amount of memory and a large bandwidth, and the algorithm is intuitive. It can be composed of logical operations that are easier than complex arithmetic operations. Thus, even today, when the computing system is rapidly developed, it is widely used in the field of extracting and recognizing objects in an image or analyzing object movement.

In order to extract and recognize an object in an image or analyze object movement using a binary image, it is important to extract information about the shape, position, and orientation of an object region, that is, the region occupied by the object in the image. It can also be extracted from the outline, which is against the background.

That is, since it is possible to extract and recognize an object or analyze object movement using only the outline of the object without using the information of the entire image, it is possible to use the system data such as memory and CPU and to process the desired image data. You can get it quickly. Therefore, in the application field based on binary image, instead of using the binary image itself for image analysis as it is, only the contour information of the object existing in the binary image is converted into a desired form.

On the other hand, when analyzing the motion of an object in a binary image, the contour of the object is represented by a chain code for rapid computational processing of the system.

The chain code represents the relationship between each pixel constituting the outline while following the outline of an object existing in the binary image. The relationship between each pixel is the pixel of the current position constituting the outline and the next in the process of searching the outline. It means a position correlation between the pixels constituting the contour to be searched by. By using the chain code, the positional correlation between each pixel corresponding to the outline starting from the starting pixel of the object and back to the starting pixel is found. In addition, since the chain code has information according to the positional correlation between each pixel corresponding to the outline of the object as described above, it is also possible to restore the original binary image from the chain code in reverse.

However, the conventional object contour chain code expression method is a method of analyzing the motion of an object by using the chain code by expressing the contour of the object existing in the image by a chain code, when analyzing the object motion in the two-dimensional image Although it is applicable to a multidimensional image, there is a problem that cannot be applied due to the multidimensional structure of an object.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and provides an apparatus and method for expressing chain motion for object motion analysis, which enables the movement of an object existing in a multidimensional image as a chain code. have.

In accordance with an aspect of the present invention, there is provided a chain code representation apparatus for object motion analysis, comprising: an image implementation unit configured to receive a captured image from a photographing apparatus and to implement a multi-dimensional image; It is preferable to include a code expression unit for analyzing the motion trajectory of the object present in the multi-dimensional image implemented by the image implementer to express the chain code.

On the other hand, the chain code representation method for object motion analysis according to an embodiment of the present invention includes the steps of detecting the position coordinates of the object that changes according to the motion trajectory of the object present in the multi-dimensional image; Setting a direction vector between the detected position coordinates of the object according to the movement trajectory of the object; The method may further include converting and arranging each direction vector set between position coordinates of an object into a chain code using a radial chain code map.

According to the apparatus and method for expressing a chain code for object motion analysis according to the present invention, by expressing the motion of an object existing in a multi-dimensional image with a chain code, the computational processing speed of the system for object motion analysis can be improved in a multi-dimensional image field. It can be effective.

Hereinafter, with reference to the accompanying drawings will be described in detail with respect to the chain code representation apparatus and method for object motion analysis according to an embodiment of the present invention.

1 is a diagram schematically illustrating a configuration of a chain code representation apparatus for analyzing object motions according to an embodiment of the present invention.

Referring to FIG. 1, the chain code representation apparatus for object motion analysis according to the present invention includes an image implementation unit 100 and a code representation unit 200.

The image implementer 100 receives a plurality of captured images from the photographing apparatus and implements a multidimensional image by applying a 3D stereo algorithm or a multi-view algorithm. Here, it is preferable that an imaging device contains a some camera etc.

The code expression unit 200 analyzes the motion trajectory of the object existing in the multi-dimensional image implemented by the image implementation unit 100 and expresses it as a chain code.

The code expression unit 200 includes an object position detection module 210, a direction vector setting module 220, and a chain code conversion module 230.

The object position detection module 210 detects the position coordinates of the object that is changed according to the movement trajectory of the object existing in the multi-dimensional image implemented by the image implementation unit 100. Remember the initial position coordinates of the object. For example, the initial position coordinate of the object preferably has a coordinate value of '0 0 0' on a three-dimensional space composed of an X axis, a Y axis, and a Z axis.

The direction vector setting module 220 sets a direction vector between position coordinates of the object detected by the object position detection module 210. In this case, the direction vector setting module 220 preferably sets the direction vector in the direction in which the object moves, using the initial position coordinates of the object as a starting point. That is, the direction vector represents the position change amount of the object that changes according to the movement of the object.

For example, the direction vector setting module 220 has a position coordinate of the object detected by the object position detecting module 210 as' 0 0 0, 1 0 1, 4 1 1, 6 1 0, 8 1 -1, 10 1 -3, 11 2 -6, 12 3 -9, 10 4 -11, 8 4 -14, 6 5 -15, 0 6 -14, -2 6 -12, -2 5 -9, -2 4 In the case of -7, -2 4 -7, -2 3 -5, -1 2 -3, 0 2 -1 ', it is preferable to set the direction vector between the position coordinates as shown in FIG.

The chain code conversion module 230 converts each direction vector set between the position coordinates of the object through the direction vector setting module 220 into a chain code using the radial chain code map.

The chain code conversion module 230 converts each direction vector set according to the direction of movement of the object through the direction vector setting module 220 into chain codes in order, and arranges initial position coordinates of the object. Therefore, it is preferable to arrange the converted chain codes based on the direction of the object moving from the initial position coordinate of the object.

Here, the radial chain code map preferably includes information about chain codes matched for each reference direction vector extending in all directions from the center point while having a directionality due to an angle difference of a certain size in three-dimensional space.

3 is a diagram illustrating an example of a radial chain code map according to the present invention. Referring to FIG. 3, a radial chain code map has a directionality due to an angle difference of a predetermined size from each other in a three-dimensional space. It may include information about chain codes matched for each reference direction vector extending in 26 directions.

That is, when the chain code conversion module 230 converts each direction vector set through the direction vector setting module 220 into a chain code, the chain code conversion module 230 maps each direction vector to a radial chain code map to correspond to the reference direction corresponding to each direction vector. It is desirable to search the vector and transform the respective directional vectors into chain codes that match the retrieved reference direction vectors. Herein, when the chain code conversion module 230 searches for a reference direction vector corresponding to each direction vector, it is preferable to search for a reference direction vector having a direction closest to the direction of each direction vector using an approximation algorithm. desirable.

For example, the chain code conversion module 230 maps each direction vector between the position coordinates set as shown in FIG. 2 through the direction vector setting module 220 to the radial chain code map shown in FIG. When converted to a code, the chain code array is of the form '0 0 0 10 2 10 19 19 20 20 22 23 22 4 14 14 16 16 16 20 10'.

On the other hand, the chain code is preferably composed of numbers, alphabets or a combination of numbers and alphabets. For example, each direction vector between position coordinates set as shown in FIG. 2 by replacing alphabets of 10 or more may be represented by a chain code character array having a form of '0 0 0 A 2 A J J K K M N M 4 E E G G G K A'.

FIG. 4 is a view sequentially illustrating a chain code representation method for object motion analysis according to an embodiment of the present invention. Referring to FIG. 4, a chain code representation method for object motion analysis according to the present invention will be described. As follows.

First, the code expression unit 200 detects position coordinates of an object that changes according to a motion trajectory of an object existing in the multi-dimensional image implemented through the image implementation unit 100 (S300).

The code expression unit 200 in the above step S300 While detecting the position coordinates of the object that changes according to the movement trajectory of the object, It is desirable to remember the initial position coordinates of the object.

After the above step S300, the code expression unit 200 sets the direction vector between the position coordinates of the object detected by the object position detection module 210 according to the movement trajectory of the object (S310).

In the above-described step S310, the code expression unit 200 may set the direction vector in the direction in which the object moves by using the initial position coordinates of the object as a starting point.

After the above step S310, the code expression unit 200 converts each direction vector set between the position coordinates of the object into a chain code by using the radial chain code map (S320). Here, the radial chain code map preferably includes information about chain codes matched for each reference direction vector extending in all directions from the center point while having a directionality due to an angle difference of a certain size in three-dimensional space.

In the above-described step S320, the code converting unit converts each direction vector set according to the movement direction of the object into chain codes in order, arranges the initial position coordinates of the object, and then uses the initial position coordinates of the object as a starting point. It is preferable to arrange the corresponding converted chain codes based on the direction in which.

In the step S320, when the code expression unit 200 converts each direction vector into a chain code, the code expression unit 200 maps each direction vector to a radial chain code map to retrieve a reference direction vector corresponding to the direction of each direction vector. It is preferable to convert the respective directional vectors into chain codes matching the retrieved reference direction vectors.

The apparatus and method for expressing a chain code for object motion analysis according to the present invention can be variously modified and implemented within the scope of the technical idea of the present invention without being limited to the above-described embodiment.

1 is a diagram schematically illustrating a configuration of a chain code representation apparatus for analyzing object motions according to an embodiment of the present invention.

2 is a diagram illustrating a direction vector set between position coordinates of an object by the direction vector setting module according to the present invention.

3 shows an exemplary radial chain codemap in accordance with the present invention.

4 is a diagram sequentially illustrating a chain code representation method for analyzing object motions according to an embodiment of the present invention.

*** Explanation of symbols for the main parts of the drawing ***

100: image implementation unit 200: code expression unit

210: object position detection module 220: direction vector setting module

230: chaincode conversion module

Claims (15)

An image realization unit configured to receive a captured image from a photographing apparatus and implement a multi-dimensional image; And a code expressing unit for analyzing a motion trajectory of an object existing in the multi-dimensional image implemented by the image realizing unit and expressing the chain trace as a chain code. The method of claim 1, The image implementer, A chain code representation apparatus for object motion analysis, comprising: receiving a captured image from the photographing apparatus and applying a 3D stereo algorithm or a multi-view algorithm to implement a multidimensional image. The method of claim 1, The code expression unit, An object position detection module detecting position coordinates of an object that changes according to a motion trajectory of an object existing in the multi-dimensional image implemented by the image implementer; A direction vector setting module for setting a direction vector between position coordinates of the object detected by the object position detection module; And a chain code conversion module for converting and arranging each direction vector set between the position coordinates of the object through the direction vector setting module using a radial chain code map. Code representation device. The method of claim 3, The object position detection module, While detecting the position coordinates of the object that changes according to the movement trajectory of the object, A chain code representation apparatus for analyzing object motions, comprising: storing initial position coordinates of an object. The method of claim 4, wherein The chain code conversion module, Each direction vector set according to the direction of movement of the object is sequentially converted into a chain code through the direction vector setting module, and the initial position coordinates of the object stored in the object position detection module are arranged, A chain code representation apparatus for object motion analysis, wherein the converted chain codes are arranged based on the direction in which the object moves from the position coordinates. The method according to any one of claims 3 to 5, The radial chain code map, Chain code for object motion analysis, which includes information about chain codes matched for each reference direction vector extending in all directions from the center point while having a direction by a certain degree of angular difference between each other in three-dimensional space Presentation device. The method of claim 6, The chain code conversion module, When converting each direction vector set through the direction vector setting module into a chain code, each direction vector is mapped to a radial chain code map to search for a reference direction vector corresponding to each direction vector, and match the searched reference direction vector. And a chain code for converting each direction vector set by the direction vector setting module into a chain code. The method of claim 7, wherein The chain cord is, A chain code representation apparatus for object motion analysis, characterized in that consisting of numbers, alphabets or a combination of numbers and alphabets. Detecting a position coordinate of an object that changes according to a motion trajectory of the object existing in the multidimensional image; Setting a direction vector between the detected position coordinates of the object according to the movement trajectory of the object; And converting and arranging each direction vector set between position coordinates of an object into a chain code using a radial chain code map. 10. The method of claim 9, Detecting the position coordinates of the object, And storing the initial position coordinates of the object. The method of claim 10, Setting the direction vector, And a direction vector is set in a direction in which the object moves using the object initial position coordinates as a starting point. The method of claim 10, Arranging the direction vectors by converting them into chain codes, Each direction vector set according to the movement direction of the object is converted into a chain code, and arranged. The initial position coordinates of the object are arranged, and then the converted chain is based on the direction in which the object moves from the initial position coordinates of the object to the starting point. A chain code representation method for object motion analysis, characterized by arranging codes. The method according to any one of claims 9 to 12, The radial chain code map, Chain code for object motion analysis, which includes information about chain codes matched for each reference direction vector extending in all directions from the center point while having a direction by a certain degree of angular difference between each other in three-dimensional space Express way. The method of claim 13, Arranging the direction vectors by converting them into chain codes, Mapping each direction vector to a radial chain codemap to retrieve a reference direction vector corresponding to the direction of each direction vector; And converting each direction vector into a chain code matched with the retrieved reference direction vector. The method of claim 14, The chain cord is, A chain code representation method for object movement analysis, characterized by consisting of numbers, alphabets, or a combination of numbers and alphabets.
KR1020090020204A 2009-03-10 2009-03-10 Chain code representation method and apparatus for object motion analysis KR20100101812A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020090020204A KR20100101812A (en) 2009-03-10 2009-03-10 Chain code representation method and apparatus for object motion analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020090020204A KR20100101812A (en) 2009-03-10 2009-03-10 Chain code representation method and apparatus for object motion analysis

Publications (1)

Publication Number Publication Date
KR20100101812A true KR20100101812A (en) 2010-09-20

Family

ID=43007141

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020090020204A KR20100101812A (en) 2009-03-10 2009-03-10 Chain code representation method and apparatus for object motion analysis

Country Status (1)

Country Link
KR (1) KR20100101812A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120062169A (en) * 2010-12-06 2012-06-14 한국전자통신연구원 Apparatus and method for incoding by using depth chaincode
KR101326029B1 (en) * 2011-03-04 2013-11-07 (주)나임기술 Method for image processing in sorter

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120062169A (en) * 2010-12-06 2012-06-14 한국전자통신연구원 Apparatus and method for incoding by using depth chaincode
KR101326029B1 (en) * 2011-03-04 2013-11-07 (주)나임기술 Method for image processing in sorter

Similar Documents

Publication Publication Date Title
JP6858650B2 (en) Image registration method and system
US9569695B2 (en) Adaptive search window control for visual search
KR101399804B1 (en) Method and apparatus for tracking and recognition with rotation invariant feature descriptors
JP5940453B2 (en) Method, computer program, and apparatus for hybrid tracking of real-time representations of objects in a sequence of images
US10554957B2 (en) Learning-based matching for active stereo systems
WO2017095576A1 (en) Method and system of curved object recognition using image matching for image processing
CN112990228B (en) Image feature matching method, related device, equipment and storage medium
US9524432B1 (en) Fast interest point extraction for augmented reality
KR101913336B1 (en) Mobile apparatus and method for controlling the same
US20120099796A1 (en) Image patch descriptors
JP2018151748A (en) Image processing device, image processing method, template generation device, object recognition processing device, and program
KR20140102038A (en) Video matching device and video matching method
CN104156998A (en) Implementation method and system based on fusion of virtual image contents and real scene
JP2014041486A (en) Signal processing method and signal processing apparatus
JP2022179397A (en) Depth estimation method for variable focus camera
JP5477518B1 (en) Image processing apparatus, image processing method, and image processing program
KR101586071B1 (en) Apparatus for providing marker-less augmented reality service and photographing postion estimating method therefor
JP5536124B2 (en) Image processing system and image processing method
CN112465796B (en) Light field feature extraction method integrating focal stack and full-focus image
CN108109164B (en) Information processing method and electronic equipment
JP5500404B1 (en) Image processing apparatus and program thereof
JP6016242B2 (en) Viewpoint estimation apparatus and classifier learning method thereof
KR20100101812A (en) Chain code representation method and apparatus for object motion analysis
CN112750164B (en) Lightweight positioning model construction method, positioning method and electronic equipment
CN115482285A (en) Image alignment method, device, equipment and storage medium

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E601 Decision to refuse application