KR101732402B1 - Gait monitoring apparatus by video observation and method of thereof - Google Patents

Gait monitoring apparatus by video observation and method of thereof Download PDF

Info

Publication number
KR101732402B1
KR101732402B1 KR1020150189010A KR20150189010A KR101732402B1 KR 101732402 B1 KR101732402 B1 KR 101732402B1 KR 1020150189010 A KR1020150189010 A KR 1020150189010A KR 20150189010 A KR20150189010 A KR 20150189010A KR 101732402 B1 KR101732402 B1 KR 101732402B1
Authority
KR
South Korea
Prior art keywords
image
gait
unit
determination unit
moving object
Prior art date
Application number
KR1020150189010A
Other languages
Korean (ko)
Inventor
김은이
유지은
최진희
Original Assignee
건국대학교 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 건국대학교 산학협력단 filed Critical 건국대학교 산학협력단
Priority to KR1020150189010A priority Critical patent/KR101732402B1/en
Application granted granted Critical
Publication of KR101732402B1 publication Critical patent/KR101732402B1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/188—Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/112—Gait analysis
    • G06K9/00624—
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19608—Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
    • H04N5/23219—

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Physiology (AREA)
  • Signal Processing (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Multimedia (AREA)
  • Dentistry (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

The present invention relates to a gait monitoring apparatus through video surveillance. The gait monitoring apparatus includes a video preprocessing unit for performing a preprocessing process for removing a background from an image captured in real time in an image acquiring unit, a moving object extracting unit for extracting moving objects from the preprocessed video frame in the image preprocessing unit, A first judging unit for judging a moving object to be in a tingling state when it is included in consecutive image frames exceeding the time, a moving unit for moving a moving object And a second determination unit for extracting and tracking a silhouette of the moving object to generate a walking image of the moving object; And a database unit for storing a gait pattern corresponding to the moving object in correspondence with the gait pattern of the moving object, the processing unit searches the gait pattern generated in the second determination unit in the gait pattern stored in the database unit and outputs a similar gait pattern . Therefore, crime can be prevented in advance through real-time image monitoring.

Description

TECHNICAL FIELD [0001] The present invention relates to a monitoring apparatus and method for monitoring gait through a video surveillance,

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to an apparatus and method for monitoring a gait through video surveillance.

It is necessary to monitor the image after the occurrence of the crime and judge the intruder by using only the face, the body shape, or the attire in the investigation of the crime that occurs by invading the residence or the shop.

On the other hand, in an example of using a video image for a crime investigation, it is judged that the person is walking and compared with the suspect who is on the investigation line. Since this is subjective and is performed after the occurrence of the crime, .

In addition, there is a need for a method to detect a person who is classified as a dangerous person in advance in the image before the event, and prevent the occurrence of the event in advance.

Korean Registered Patent No. 1148234

According to an aspect of the present invention, there is provided a method of detecting a person object, comprising: detecting a human object included in the image to determine a tingling of a human object; analyzing a gait of the person object determined to be tingling; To prevent crime in advance.

The gait monitoring apparatus according to an embodiment of the present invention includes an image preprocessing unit 210 for performing a preprocessing for removing a background from an image captured by the image acquisition unit 100 in real time, an image preprocessing unit 100 A first determination unit 220 for extracting a moving object from the preprocessed image frame and determining that the moving object is in a tingling state when the moving object is included in the consecutive image frames so that the moving object exceeds the set time, A second determination unit 220 for extracting and tracking a silhouette of the moving object in a plurality of image frames including the moving object determined to be in the tingling state in the first determination unit 220 to generate a walking image of the moving object, A processing unit 200 including a processor 230; And a database unit 300 for storing a gait pattern corresponding to the moving object so that the processing unit 200 can display a gait image generated by the second determination unit 230 in the database 300 And a similar gait pattern is output by searching in the gait pattern stored in the gait pattern.

The image preprocessing unit 210 performs noise elimination, morphology calculation, Gaussian or median filtering of the image received from the image acquiring unit 100.

The first determination unit 220 determines an object to be moved from an image frame that has been preprocessed by the image preprocessing unit 210 and determines whether the moving object is a recognition object, 221); A timeout determination unit 222 for determining whether the moving object determined as a human object by the object determination unit 221 is included in a plurality of image frames exceeding a predetermined time; And an object state determination unit (223) for determining that the moving object determined to be included in a plurality of image frames exceeds the set time in the time lapse determination unit (222) do.

The second determination unit 230 may include a silhouette processing unit 231 for extracting a silhouette from image frames continuous to the moving object determined as being in a state of being squeezed in the first determination unit 220; An object tracking unit (232) for tracking the silhouette of the moving object determined to be in a tingling state in successive image frames; A gait image generating unit (233) for generating a gait image including the average information of the silhouettes by collecting the silhouettes tracked in successive image frames; And a gait determining unit 234 for determining whether the gait image generated by the gait image generating unit 233 exists in the database unit 300.

The object tracking unit 232 sets the search radius based on the silhouette extracted from the previous image frame, sets the center direction in which the object is distributed in the current image frame, And tracking an object similar to the silhouette.

The object tracking unit 232 tracks the silhouette using a histogram similarity measurement algorithm.

The database unit 300 stores a gait pattern corresponding to the object information designated as the dangerous phase or the safety phase and stores the gait pattern corresponding to the new object information and the gait pattern for the new object information from the second determination unit 230. [ Is received, and is stored and updated.

If the gait generation image generated by the second determination unit 230 is searched in the database unit 300 and the gait pattern corresponding to the detected object information corresponds to the object information designated as a dangerous level, And the notification unit 240 further comprises a notification unit 240. FIG.

The second determination unit 230 searches the generated gait image in the database unit 300 and outputs the gait patterns stored in the database unit 300 similar to the gait images in order of high similarity .

According to an embodiment of the present invention, a method for monitoring a gait through video surveillance includes: (a) acquiring an image; (b) performing a preprocessing operation to delete the background image of the acquired image; (c) determining a state of an object by distinguishing a human object from an image and determining whether or not a time is exceeded; (d) distinguishing a human object from a video, detecting and tracking a silhouette to determine a gait by generating a gait image; And (e) generating a warning according to the determined object state and gait.

In the step (c), if the object state is a human object, the step (d) is performed. Otherwise, the step is terminated.

According to this aspect, an apparatus and method for monitoring a gait through video surveillance according to an embodiment of the present invention detects a human object in an image and regards the human object as a surge when it is included in an image for a predetermined time or more, A person who is regarded as a distance is traced and a person corresponding to the stepped gauges is formed into a database through the generation and discrimination of a gait image, thereby preventing the crime through monitoring the image captured in real time.

1 is a block diagram showing a schematic configuration of a gait monitoring apparatus according to an embodiment of the present invention.
2 is a block diagram showing a schematic configuration of a first determination unit of a gait monitoring apparatus according to an embodiment of the present invention.
3 is a block diagram illustrating a schematic configuration of a second determination unit of a gait monitoring apparatus according to an embodiment of the present invention.
4 is a flowchart illustrating a gait monitoring method using video surveillance according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating object tracking in a second determination unit of the gait monitoring apparatus according to an embodiment of the present invention.
FIG. 6 is a diagram illustrating object tracking in the second determination unit of the gait monitoring apparatus according to the embodiment of the present invention.
FIG. 7 is a view illustrating an output result of a gait monitoring apparatus according to an embodiment of the present invention.
8 is a view showing an output result of a gait monitoring apparatus according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Hereinafter, an apparatus and method for monitoring a gait by video surveillance according to an embodiment of the present invention will be described with reference to the accompanying drawings.

1, a gait monitoring apparatus according to an embodiment of the present invention will be described with reference to FIGS. 1 to 3. FIG. 1 is a block diagram illustrating a gait monitoring apparatus according to an embodiment of the present invention. A processing unit 200 and a database unit 300. The processing unit 200 includes an image preprocessing unit 210, a first determination unit 220, a second determination unit 230, and a notification unit 240, .

The image acquisition unit 100 acquires an image in real time from a camera installed in a monitoring target place, and transmits the acquired image to the processing unit 200 through a wired or wireless connection.

The processing unit 200 receives an image from the image acquiring unit 100 in a wired or wireless connection with the image acquiring unit 100, and performs a preprocessing in the image preprocessing unit 210.

In one example, the image preprocessing unit 210 removes the background by separating the background and the image from the image received from the image acquiring unit 100, compares the consecutive frames, and outputs the difference between the image of the current frame and the pixel difference To perform motion detection.

At this time, the image preprocessing unit 210 may perform pre-processing such as noise elimination, morphology calculation, Gaussian filtering, or median filtering of the image received from the image obtaining unit 100.

The first determination unit 220 detects a human object from the image preprocessed by the image preprocessing unit 210, and determines whether the detected human object is tingling.

2, the first determination unit 220 includes an object determination unit 221, a timeout determination unit 222, and an object state determination unit 223. The object determining unit 221 determines the human object from the preprocessed image received from the image preprocessing unit 210 by the first determination unit 220.

At this time, the object determining unit 221 determines moving objects from continuous frames of the preprocessed image, and determines whether the discriminated object is a human object.

For example, when the human object is identified in the object determination unit 221, the timeout determination unit 222 operates. However, if the human object is not identified and the moving object is an object, the timeout determination unit 222 And performs the object determining unit 221 in the next frame.

The object discrimination unit 221 can discriminate all the moving objects included in the image frame, for example, can discriminate a plurality of moving object objects. In the description of the present invention, for example, a single moving object object is discriminated I will explain it.

The time-out determining unit 222 counts a time when the human object identified by the object determining unit 221 appears on the image frame, and if the human object exists in the image frame beyond the set time, .

More specifically, the timeout determination unit 222 determines that the corresponding human object starts to appear in the frame of the image and reaches a continuous image frame from the corresponding frame determined as a human object. In a preferred example, the timeout determination unit 222 determines that the person object is over time if the person object is present on the frame of the image for 15 seconds.

At this time, the time-out determining unit 222 transmits the information of the person object determined to be over time to the object state determining unit 223, and the object state determining unit 223 notifies the state of the person object, It is judged to be loitering.

At this time, the object state determining unit 223 transmits the determined person object information to the second determining unit 230, and determines whether the continuous image frame monitored for the timeout reference time (for example, 15 seconds) To the second judging unit 230. [0050]

3, the second determination unit 230 includes a silhouette processing unit 231, an object tracking unit 232, a gait generation image generation unit 233, and a gait determination unit 234, And determines the stepping of the person object from the continuous image frame transmitted from the unit 220.

More specifically, the silhouette processing unit 231 extracts a silhouette of a human object from successive image frames for a human object determined to be in a state of being tumbled by the first determination unit 220, Tracks the silhouette object extracted from the silhouette processing unit 231.

At this time, the object tracking unit 232 sets the search radius based on the current position as shown in FIG. 5, and as it proceeds from the current position toward the distribution center in the arrow direction, the object being tracked, Estimate similar objects. Accordingly, the object tracking unit 232 can track a person object that is stationary or moving in a plurality of consecutive image frames.

At this time, the object tracking unit 232 tracks the human object according to the similarity degree of the human object from a plurality of consecutive image frames using the histogram similarity measurement algorithm of FIG.

In one example, the object tracking unit 232 transmits the result of silhouette tracking of the human object over a plurality of consecutive image frames to the gait image generating unit 233.

Accordingly, the gait image generation unit 233 generates a gait image using the result of silhouette tracking of the human object with respect to a plurality of consecutive image frames received from the object tracking unit 232.

For example, the gait image generation unit 233 collects 30 image frames received from the object tracking unit 232 and generates a gait energy image (GEI) including average information of the average silhouette tracking result do.

At this time, the gait image generated by the gait image generating unit 233 includes a specific gait pattern of the human object including the result of the silhouette tracking.

At this time, the gait determining unit 234 searches the database 300 for the gait image generated by the gait image generating unit 233, and determines whether or not the gait image is present in the database 300.

In one example, when the gait determination unit 234 searches the gait image in the database 300, if the database 300 does not store the gait target gait, the gait determination unit 234 determines And stores the gait image generated by the gait image generating unit 233 in the database 300.

In another example, when the gait detection unit 234 searches for the gait image in the database 300, if the gait image is not present in the database 300, the gait detection unit 234 determines that the gait And stores the gait generation image generated by the image generation unit 233 in the database 300 as new data.

3, the second determination unit 230 extracts a silhouette of a human object from a plurality of image frames, generates a gait image by tracking the silhouette of the corresponding human object, And searches the database 300 for an image of the gait.

In this case, the first determination unit 220 is described based on identification of a single human object, and the second determination unit 230 has been described as generating a gait image for a single human object , The first determination unit 220 determines a timeout for a plurality of human objects belonging to the image frame and the second determination unit 230 generates a gait image for each human object It should be interpreted as being.

Referring to FIG. 1 again, the processing unit 200 of the gait monitoring apparatus according to an embodiment of the present invention will be described. The notification unit 240 receives a gait detection signal from the gait detection unit 230 of the second determination unit 230, (Step 234), i.e., whether or not the gait pattern is present in the database unit 300. [0050]

For example, the notification unit 240 determines that a gait pattern corresponding to the gait image generated by the second determination unit 230 is present in the database unit 300, and that a person object corresponding to the gait pattern exists in the database unit 300), it generates a notification of the dangerous situation.

At this time, the notification unit 240 can notify the generated notification to the police or the security-related organization. In one example, the notification unit 240 may transmit a notification to a related organization through a separate communication module.

The database unit 300 stores a gait pattern and object information corresponding to the pattern. In one example, the database unit 300 stores a gait pattern and object information in a one-to-one correspondence relationship.

At this time, the object information corresponding to the gait pattern may be a person object designated as a dangerous person, for example, a criminal or a suspect, or a person object designated as a safe person. In the case of a person object designated as a safe person, it may be a resident of a space that is the object of video surveillance.

In one example, if the video surveillance target is a residential space, the resident of the space to be surveilled is designated as a safe person. The database unit 300 displays a gait pattern for the human objects of the family members residing in the residential space . Accordingly, even if the gait pattern of the human object with respect to the family generated by the second determination unit 230 exists in the database unit 300, a notification is generated for the person object for the corresponding gait pattern designated as a safe person I never do that.

In the case where the video surveillance target is a business site such as a shop, the related persons such as a president of a business site or a part-time student are designated as a safe person, and the database unit 300 includes a gait pattern of the persons concerned. Accordingly, even if the second determination unit 230 requests the database unit 300 to search for the gait pattern of the corresponding person object, and the gait pattern exists in the database unit 300, the second determination unit 230 The person object for the gait pattern is not classified as a dangerous state, and no notification is generated.

The gait monitoring apparatus according to an embodiment of the present invention is configured to have a structure as described with reference to Figs. 1 to 3, so that a moving person object is extracted from a captured image in real time, If the image is included in the image longer than the set time, the gait pattern is generated through the extraction of the silhouette of the human object and the gait pattern is searched in the database unit 300 to determine whether the person object is dangerous, It is possible to detect suspicious person such as sagging, and it is effective in crime prevention.

Next, a gait monitoring method using video surveillance through a gait monitoring apparatus using video surveillance according to an embodiment of the present invention will be described in detail with reference to the flow chart of FIG. 4 and the output screen of FIG. 7 and FIG.

As shown in FIG. 4, the gait monitoring method using video surveillance according to an embodiment of the present invention includes steps of acquiring an image (S100), performing a preprocessing operation to delete a background image of an acquired image (S210) (S220) of determining a state of an object by distinguishing a human object from the image and determining whether the time is over (S220), determining a gait by dividing a human object in the image and detecting and tracking a silhouette to generate a gait image S230) and generating a warning according to the determined object state and gait (S240).

First, in step S100, an image is acquired by the image acquisition unit 100 of the gait monitoring apparatus through video surveillance.

Next, the preprocessing operation (S210) for deleting the background image of the acquired image is performed by the image preprocessing unit 210 among the processing unit 200 of the gait monitoring apparatus through the video surveillance.

In operation S220, the first determination unit 220 of the processing unit 200 of the gait monitoring apparatus according to the present invention monitors the object state, 7, when the human object detected in the image is displayed on the image after exceeding the set time, the state of the corresponding human object is tilted as shown in the second output screen of FIG. 7 .

In one example, if the human object is included in the image in a range not exceeding the set time in step S220, the process ends without executing the following steps.

However, if the human object is detected in the image in a time longer than the set time in step S220, the step of discriminating the human object from the image to be executed and detecting and tracking the silhouette to determine the stepping through the generation of the stepping image S230) is performed in the second determination unit 230 of the processing unit 200 of the gait monitoring apparatus through video surveillance.

In step S230, the gait image is formed by tracking the silhouette of the human object included in the image as shown in the third output screen of Fig. 7, and the gait generation image is searched in the database unit 300. [ At this time, out of the gait patterns stored in the database unit 300, a gait pattern similar to the search target gait image is output in a similar order as the first output screen of Fig.

As such, it is possible to confirm the result of searching the database unit 300 for a gait pattern similar to the gait image in the order of high similarity.

At this time, as shown in the first output screen of FIG. 8, a person object having a gait pattern similar to a gait pattern of a detected person object in a form of outputting search results of the database unit 300 in descending order of degree of similarity Information can be output.

However, in another example, when a gait pattern having a degree of similarity to a gait pattern of a detected human object is very high, for example, 90% or more of similarity exists in the database unit 300, A step S240 of extracting information of the person object corresponding to the gait pattern from the database unit 300 and generating a warning according to the judged object state and gait to output as a notification message is performed.

For example, the degree of similarity for extracting the object information may be changed to a set value other than 90% in the step of generating the warning (S240).

As described above, the gait monitoring method using the video surveillance according to the embodiment of the present invention described with reference to the flowchart of FIG. 4 and FIGS. 7 and 8, Even when the object is wearing other clothes, it is possible to extract the silhouette of the person object, generate a gait image, and compare the gait pattern with the gait pattern stored in the database unit 300. [

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

100: image acquisition unit 200:
210: image preprocessing unit 220: first determination unit
230: second determination unit 240:
300:

Claims (11)

An image preprocessing unit 210 for performing a preprocessing process for removing a background from an image captured by the image acquiring unit 100 in real time, an image preprocessing unit 210 for extracting a moving object from the preprocessed image frame, (220) for determining that the moving object is in a tingling state when the moving object is included in the consecutive image frames exceeding the set time, A processing unit (200) comprising a second determination unit (230) for extracting and tracking a silhouette of the moving object in a plurality of image frames including the object to generate a stepping image of the moving object; And
A database unit 300 for storing a gait pattern corresponding to the moving object;
The processing unit 200 searches for the gait image generated by the second determination unit 230 in the gait pattern stored in the database unit 300 and outputs a similar gait pattern, 2, if the gait image generated by the gauging unit 230 is not stored in the database unit 300, the generated gait image is stored in the database unit 300. [ Monitoring device. The method according to claim 1,
Wherein the image preprocessing unit 210 performs noise removal, morphology calculation, Gaussian or intermediate value filtering of the image received from the image obtaining unit 100. [ The method according to claim 1,
The first determination unit 220 determines,
An object discrimination unit 221 for discriminating a moving object from an image frame that has been preprocessed by the image preprocessing unit 210 and discriminating whether the moving object discriminated as a human object is a recognition object;
A timeout determination unit 222 for determining whether the moving object determined as a human object by the object determination unit 221 is included in a plurality of image frames exceeding a predetermined time; And
An object state determination unit (223) for determining, in the timeout determination unit (222), the moving object determined to be included in a plurality of image frames exceeding the set time as a tingling state;
And monitoring the gait by the video surveillance. The method according to claim 1,
The second determination unit (230)
A silhouette processing unit (231) for extracting a silhouette from consecutive image frames of the moving object determined as a state of being tumbled in the first determination unit (220);
An object tracking unit (232) for tracking the silhouette of the moving object determined to be in a tingling state in successive image frames;
A gait image generating unit (233) for generating a gait image including the average information of the silhouettes by collecting the silhouettes tracked in successive image frames; And
A gait determining unit 234 for determining whether the gait image generated by the gait image generating unit 233 exists in the database unit 300;
And monitoring the gait by the video surveillance. 5. The method of claim 4,
The object tracking unit 232 sets the search radius based on the silhouette extracted from the previous image frame, sets the center direction in which the object is distributed in the current image frame, And tracking the object similar to the silhouette to monitor the walking gait through the video surveillance. 5. The method of claim 4,
Wherein the object tracking unit (232) tracks the silhouette using a histogram similarity measurement algorithm. The method according to claim 1,
The database unit 300 stores a gait pattern corresponding to the object information designated as the dangerous phase or the safety phase and stores the gait pattern corresponding to the new object information and the gait pattern for the new object information from the second determination unit 230. [ And a step of monitoring the gait by means of video surveillance. 8. The method of claim 7,
If the gait generation image generated by the second determination unit 230 is searched in the database unit 300 and the gait pattern corresponding to the detected object information corresponds to the object information designated as a dangerous level, And the notification unit 240 further comprises a notification unit 240 for informing the user of the gait. The method according to claim 1,
The second determination unit 230 searches the generated gait image in the database unit 300 and outputs the gait patterns stored in the database unit 300 similar to the gait images in order of high similarity Wherein the gait monitoring apparatus comprises: (a) acquiring an image;
(b) performing a preprocessing operation to delete the background image of the acquired image;
(c) determining a state of an object by distinguishing a human object from an image and determining whether or not a time is exceeded;
(d) generating a gait image by classifying the human object in the image, detecting and tracking the silhouette, searching the generated gait image in the gait pattern stored in the database, and outputting a similar gait pattern, Storing the generated gait image in the database unit when the gait image is not stored in the database unit, and determining the generated gait image; And
(e) generating an alert based on the determined object state and gait;
And monitoring the gait through the video surveillance. 11. The method of claim 10,
Wherein the step (d) is performed when the object state is determined to be a human object, and if not, the step is terminated.
KR1020150189010A 2015-12-29 2015-12-29 Gait monitoring apparatus by video observation and method of thereof KR101732402B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020150189010A KR101732402B1 (en) 2015-12-29 2015-12-29 Gait monitoring apparatus by video observation and method of thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020150189010A KR101732402B1 (en) 2015-12-29 2015-12-29 Gait monitoring apparatus by video observation and method of thereof

Publications (1)

Publication Number Publication Date
KR101732402B1 true KR101732402B1 (en) 2017-05-08

Family

ID=60164023

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020150189010A KR101732402B1 (en) 2015-12-29 2015-12-29 Gait monitoring apparatus by video observation and method of thereof

Country Status (1)

Country Link
KR (1) KR101732402B1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190081350A (en) 2017-12-29 2019-07-09 (주)동아금속 Monitoring system of walking balance for lower limb rehabilitation
CN111476198A (en) * 2020-04-24 2020-07-31 广西安良科技有限公司 Gait recognition method, device and system based on artificial intelligence, storage medium and server

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101470315B1 (en) * 2014-08-11 2014-12-09 (주)엔토스정보통신 Closed-circuit television system of sensing risk by moving of object and method thereof

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101470315B1 (en) * 2014-08-11 2014-12-09 (주)엔토스정보통신 Closed-circuit television system of sensing risk by moving of object and method thereof

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190081350A (en) 2017-12-29 2019-07-09 (주)동아금속 Monitoring system of walking balance for lower limb rehabilitation
CN111476198A (en) * 2020-04-24 2020-07-31 广西安良科技有限公司 Gait recognition method, device and system based on artificial intelligence, storage medium and server
CN111476198B (en) * 2020-04-24 2023-09-26 广西安良科技有限公司 Gait recognition method, device, system, storage medium and server based on artificial intelligence

Similar Documents

Publication Publication Date Title
JP6241796B2 (en) Left item monitoring device, left item monitoring system including the same, and left item monitoring method
KR101724658B1 (en) Human detecting apparatus and method
US7961953B2 (en) Image monitoring system
JP6764481B2 (en) Monitoring device
JP6390860B2 (en) Left object monitoring device, left object monitoring system including the same, and left object monitoring method
KR101541272B1 (en) Apparatus and method for detecting terrorism using irregular motion of peoples
JP2018042270A (en) Left object monitoring device, left object monitoring system with the same, and left object monitoring method
KR101858396B1 (en) Intelligent intrusion detection system
WO2018179202A1 (en) Information processing device, control method, and program
KR101372860B1 (en) System for searching video and server for analysing video
JP5358851B2 (en) Suspicious behavior detection method and suspicious behavior detection device
CN113887445A (en) Method and system for identifying standing and loitering behaviors in video
KR101646000B1 (en) Surveillance System And Method To Object And Area
EP2000998B1 (en) Flame detecting method and device
KR101732402B1 (en) Gait monitoring apparatus by video observation and method of thereof
JP5758165B2 (en) Article detection device and stationary person detection device
JP5752977B2 (en) Image monitoring device
JP2017102719A (en) Flame detection device and flame detection method
CN113537165B (en) Detection method and system for pedestrian alarm
EP1405279A1 (en) Vision based method and apparatus for detecting an event requiring assistance or documentation
KR101539944B1 (en) Object identification method
KR101436283B1 (en) The predicting system for anti-crime through analyzing server of images
KR20140037354A (en) System for abandoned and stolen object detection
JPH0816234A (en) Device for detecting abnormality of plant
JP5167163B2 (en) Smoke detector

Legal Events

Date Code Title Description
E701 Decision to grant or registration of patent right
GRNT Written decision to grant