US20170011529A1 - Video analysis system - Google Patents

Video analysis system Download PDF

Info

Publication number
US20170011529A1
US20170011529A1 US15/117,814 US201515117814A US2017011529A1 US 20170011529 A1 US20170011529 A1 US 20170011529A1 US 201515117814 A US201515117814 A US 201515117814A US 2017011529 A1 US2017011529 A1 US 2017011529A1
Authority
US
United States
Prior art keywords
moving object
threshold
environment information
video data
processor
Prior art date
Legal status (The legal status 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 status listed.)
Abandoned
Application number
US15/117,814
Other languages
English (en)
Inventor
Keiichi Urashita
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NEC Corp
Original Assignee
NEC Corp
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 NEC Corp filed Critical NEC Corp
Assigned to NEC CORPORATION reassignment NEC CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: URASHITA, KEIICHI
Publication of US20170011529A1 publication Critical patent/US20170011529A1/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • G06T7/2053—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G06K9/00771—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/001—
    • G06T7/2006—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/254—Analysis of motion involving subtraction of images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20004—Adaptive image processing
    • G06T2207/20008—Globally adaptive
    • G06T2207/20144—
    • G06T2207/20148—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding

Definitions

  • the present invention relates to a video analysis system, a video analysis device, a video analysis method, and a program
  • the image correction system includes an information acquisition part which acquires state information showing a state in which a photograph has been shot, a correction parameter selection part which selects a plurality of correction parameters based on the state information, and an image quality correction part which executes an image data correction process by using the selected correction parameters.
  • an information acquisition part which acquires state information showing a state in which a photograph has been shot
  • a correction parameter selection part which selects a plurality of correction parameters based on the state information
  • an image quality correction part which executes an image data correction process by using the selected correction parameters.
  • a monitoring camera which acquires video data may be installed outdoors, for example, on the street or around the entrance.
  • the environment of a place where the monitoring camera is installed change under the influence of change of the outdoor environment.
  • the environment of the installation place greatly varies depending on, for example, whether it is fine or rainy, whether it is in the daytime or at night, and whether the wind is blowing or not.
  • video data acquired by the monitoring camera changes. For example, the shadow of a moving object is clearly shown on a fine day, but the shadow is not shown on a cloudy day. On a windy day, trees or the like on the background largely swings.
  • a video analysis system has a problem that change of the environment of installation of a monitoring camera and change of video data may affect the accuracy of detection of a moving object.
  • an object of the present invention is to provide a video analysis system which solves the problem that change of the environment of installation of a monitoring camera and change of video data may affect the accuracy of detection of a moving object.
  • a video analysis system as an aspect of the present invention includes:
  • a moving object detecting means for detecting a moving object from video data by using a moving object detection parameter, the moving object moving in the video data, the video data having been acquired by the video data acquiring means, the moving object detection parameter being a parameter for detecting a moving object;
  • an environment information collecting means for collecting environment information representing an external environment of a place where the video data acquiring means is installed
  • a parameter changing means for changing the moving object detection parameter on a basis of the environment information collected by the environment information collecting means, the moving object detection parameter being used when the moving object detecting means detects the moving object.
  • a video analysis device as another aspect of the present invention includes:
  • a video data acquisition part configured to acquire video data from an external device, the video data having been acquired by the external device
  • a moving object detection part configured to detect a moving object from video data by using a moving object detection parameter, the moving object moving in the video data, the video data having been acquired by the video data acquisition part, the moving object detection parameter being a parameter for detecting a moving object;
  • an environment information collection part configured to collect environment information representing an external environment of a place where the external device having acquired the video data is installed
  • a parameter changing part configured to change the moving object detection parameter on a basis of the environment information collected by the environment information collection part, the moving object detection parameter being used when the moving object detection part detects the moving object.
  • a video analysis method as another aspect of the present invention includes:
  • the environment information representing an external environment of a place where an external device acquiring video data is installed, the moving object detection parameter being used when detecting a moving object from the video data;
  • a computer program including instructions for causing a video analysis device to function as:
  • a video data acquisition part configured to acquire video data from an external device, the video data being acquired by the external device
  • a moving object detection part configured to detect a moving object from video data by using a moving object detection parameter, the moving object moving in the video data, the video data being acquired by the video data acquisition part, the moving object detection parameter being a parameter for detecting a moving object;
  • an environment information collection part configured to collect environment information representing an external environment of a place where the external device having acquired the video data is installed
  • a parameter changing part configured to change the moving object detection parameter on a basis of the environment information acquired by the environment information collection part, the moving object detection parameter being used when the moving object detection part detects the moving object.
  • the present invention enables provision of a video analysis system in which change of the environment of installation of a monitoring camera or change of video data do not affect the accuracy of detection of a moving object.
  • FIG. 1 is a view showing the configuration overview of a whole video analysis system according to a first exemplary embodiment of the present invention
  • FIG. 2 is a block diagram showing the configuration of the video analysis system according to the first exemplary embodiment of the present invention
  • FIG. 3 is a block diagram showing an example of the configuration of an environment information collecting means included by the video analysis system according to the first exemplary embodiment of the present invention
  • FIG. 4 is a view for describing an example of a moving object detection parameter changed by a parameter changing means of a video analysis device shown in FIG. 2 ;
  • FIG. 5 is a view for describing an example of the moving object detection parameter changed by the parameter changing means of the video analysis device shown in FIG. 2 ;
  • FIG. 6 is a view for describing an example of the moving object detection parameter changed by the parameter changing means of the video analysis device shown in FIG. 2 ;
  • FIG. 7 is a view for describing an example of the moving object detection parameter changed by the parameter changing means of the video analysis device shown in FIG. 2 ;
  • FIG. 8 is a view for describing an example of the moving object detection parameter changed by the parameter changing means of the video analysis device shown in FIG. 2 ;
  • FIG. 9 is a flowchart showing an example of a process from acquisition of video data to detection of a person in the video data by the video analysis system according to the first exemplary embodiment of the present invention.
  • FIG. 10 is a flowchart showing an example of a process for detecting a person in video data by the video analysis system according to the first exemplary embodiment of the present invention
  • FIG. 11 is a flowchart showing an example of a process for changing a moving object detection parameter based on environment information by the video analysis system according to the first exemplary embodiment of the present invention
  • FIG. 12 is a block diagram showing the configuration of a video analysis system according to a second exemplary embodiment of the present invention.
  • FIG. 13 is a flowchart showing an example of a process for changing a moving object detection parameter based on environment information by the video analysis system according to the second exemplary embodiment of the present invention
  • FIG. 14 is a block diagram showing the configuration overview of a video analysis system according to a third exemplary embodiment of the present invention.
  • FIG. 15 is a block diagram showing the configuration overview of a video analysis device according to a fourth exemplary embodiment of the present invention.
  • a moving object detection parameter used when the video analysis device 4 detects a moving object in video data is properly changed based on environment information collected by an environment information collecting means 3 .
  • this exemplary embodiment will describe a video analysis system 1 which detects a moving person in video data of video analysis systems 1 which detect moving objects.
  • the video analysis system 1 has the monitoring camera 2 (the video data acquiring means), the environment information collecting means 3 , and the video analysis device 4 . Further, the video analysis device 4 has a video analyzing means 41 (a moving object detecting means), a parameter changing means 42 , and an analysis result displaying means 43 .
  • the video analysis device 4 in this exemplary embodiment is an information processing device which includes an arithmetic device and a storage device. The video analysis device 4 realizes the respective functions described above by execution of a program stored in the storage device by the arithmetic device.
  • the monitoring camera 2 is installed at a predetermined position and has a function of acquiring video data at the predetermined position where the monitoring camera 2 is installed.
  • the monitoring camera 2 is installed, for example, at the airport, at an oil plant, and on the street, and is connected to the video analysis device 4 via an external network or the like.
  • the monitoring camera 2 processes video data acquired by the monitoring camera 2 into a predetermined format, and thereafter, transmits the video data to the video analyzing means 41 included by the video analysis device 4 via the external network or the like.
  • This exemplary embodiment will describe a case where the video analysis system 1 includes one monitoring camera 2 .
  • the video analysis system 1 according to this exemplary embodiment is not limited to the case where the video analysis system 1 includes the one monitoring camera 2 .
  • the video analysis system 1 may be configured to acquire video data from a plurality of monitoring cameras 2 .
  • the video data acquiring means included by the video analysis system 1 is not limited to a case of using the monitoring camera 2 .
  • the video analysis system 1 may include any video data acquiring means which acquires video data, instead of the monitoring camera 2 .
  • the environment information collecting means 3 has a function of collecting environment information representing the external environment of a place where the monitoring camera 2 is installed. For example, the date, time (daytime, night), weather information, and latitude and longitude of a place where the monitoring camera 2 is installed are regarded as the external environment.
  • the environment information is information for grasping the external environment of a place where the monitoring camera 2 is installed (i.e., a shooting environment condition).
  • the environment information collecting means 3 has, for example, a calendar 31 , a Web information collecting means 32 , and various sensors 33 such as a GPS sensor, a heliograph, a rain gauge and an anemometer.
  • the video analysis system 1 grasps the state of the shooting environment (the external environment) in the vicinity of the monitoring camera 2 by using the environment information collecting means 3 .
  • various physical sensors such as the GPS sensor, the heliograph, the rain gauge and the anemometer need to be installed in the vicinity of the monitoring camera 2 acquiring video data.
  • the calendar 31 , the Web information collecting means 32 , and so on, that are not the physical sensors do not always need to be installed in the vicinity of the monitoring camera 2 .
  • (part of) the function as the environment information collecting means 3 may be included by the video analysis device 4 to be described later.
  • the calendar 31 is used to collect information such as the sunrise, the sunset, the moon's age, and the culmination altitude.
  • the Web information collecting means 32 is used to collect information such as weather forecast information and the presence or absence of any event on the Internet.
  • the various sensors 33 are used to grasp, for example, the time, sunshine duration, rainfall, and wind speed of a place where the monitoring camera 3 is installed.
  • the environment information collecting means 3 collects information such as the latitude and longitude, the height and the time by using the GPS sensor, for example.
  • the environment information collecting means 3 collects a sunshine duration by using a heliograph, a rainfall by using a rain gauge, and a wind speed by using an anemometer, for example.
  • the environment information collecting means 3 collects various information (environment information) for grasping the external environment of the monitoring camera 2 by using the calendar 31 , the Web information collecting means 32 and the various sensors 33 , for example. Then, the environment information collecting means 3 transmits the various environment information collected thereby to the parameter changing means 42 of the video analysis device 4 .
  • the environment information collecting means 3 may be configured to acquire environment information by using a method other than the abovementioned one.
  • the environment information collecting means 3 can include a meteorological instrument other than the abovementioned ones.
  • the environment information collecting means 3 do not need to include the calendar 31 , the Web information collecting means 32 and the various sensors 33 all.
  • the environment information collecting means 3 only needs to be configured to be capable of collecting information (environment information) necessary to grasp the external environment of a place where the monitoring camera 2 is installed.
  • the video analysis device 4 has a function of detecting a person moving in video data acquired by the monitoring camera 2 by using a moving object detection parameter.
  • the video analysis device 4 also has a function of changing a moving object detection parameter based on environment information collected by the environment information collecting means 3 .
  • the video analysis device 4 is an information processing device which includes a storage device and an arithmetic device.
  • the video analysis device 4 realizes the video analyzing means 41 , the parameter changing means 42 and the analysis result displaying means 43 by execution of a program stored in the storage device by the arithmetic device.
  • the video analyzing means 41 has a function of detecting a person moving in video data acquired by the monitoring camera 2 by using a moving object detection parameter.
  • the video analyzing means 41 also has a function of acquiring the trajectory of the detected moving person, and so on.
  • the video analyzing means 41 is connected with the monitoring camera 2 via an external network or the like and acquires video data acquired by the monitoring camera 2 from the monitoring camera 2 . Then, the video analyzing means 41 detects a person moving in the acquired video data by using a moving object detection parameter.
  • the video analyzing means 41 in this exemplary embodiment obtains a difference of image data extracted from video data, and thereby finds an unstill object.
  • the video analyzing means 41 obtains a difference between image data of a previous frame (an integration result) and image data of a current frame, and thereby finds an unstill object.
  • the video analyzing means 41 removes a predetermined object of the unstill objects as a moving object other than a detection target object (a person) from a detection target.
  • the video analyzing means 41 removes a moving object other than a person from a detection target by using a reciprocating object removal filter, a shadow removal filter or the like to be described later.
  • the video analyzing means 41 is capable of grasping what the unstill object is (for example, whether the unstill object is a person or not) by using a technique such as pattern recognition.
  • the video analyzing means 41 detects a person moving in the video data. Then, the video analyzing means 41 tracks the grasped person by using a particle filter or the like, and acquires the trajectory of movement of the detected person. Further, the video analyzing means 41 transmits the detected person and the trajectory of the movement of the detected person as the detection result to the analysis result displaying means 43 .
  • the abovementioned moving object detection parameter is used, for example, when the video analyzing means 41 obtains a difference of image data to find an unstill object, and when the video analyzing means 41 uses a filter for removing a moving object other than a person.
  • the details of the moving object detection parameter will be described later.
  • the parameter changing means 42 has a function of changing a moving object detection parameter used when the video analyzing means 41 detects a moving object in video data, by using environment information acquired from the environment information collecting means 3 .
  • the parameter changing means 42 is connected with the environment information collecting means 3 , and acquires environment information collected by the environment information collecting means 3 .
  • the parameter changing means 42 changes a moving object detection parameter based on (the external environment represented by) the acquired environment information (for example, so as to make the moving object detection parameter optimum for the external environment).
  • Change of a moving object detection parameter by the parameter changing means 42 is carried out by using a parameter set, which is a set of moving object detection parameters previously determined in accordance with a possible external environment, for example.
  • the parameter changing means 42 may be configured to change each moving object detection parameter so that the moving object detection parameter becomes the most appropriate to the external environment represented by the acquired environment information.
  • the parameter changing means 42 acquires environment information from the environment information collecting means 3 . Then, the parameter changing means 42 changes a moving object detection parameter used when the video analyzing means 41 detects a moving object in video data, based on the acquired environment information.
  • the analysis result displaying means 43 is connected to the output device such as the monitor.
  • the analysis result displaying means 43 acquires an analysis result of analysis by the video analyzing means 41 (a person, the trajectory of movement of the person, and so on, as the result of the detection) from the video analyzing means 41 .
  • the analysis result displaying means 43 then causes the output device such as the monitor to display the analysis result.
  • the video analysis system 1 (for example, the video analysis device 4 thereof) may include an analysis result accumulating means for storing the results of analysis by the video analyzing means 41 , and a video data accumulating means for storing video data.
  • the video analysis system 1 can include an operation checking means for checking a system operation using a moving object detection parameter changed by the abovementioned configuration, and a parameter regulating means for autonomously conducting fine regulation of a moving object detection parameter in response to the check of the operation by the operation checking means.
  • a moving object detection parameter is a parameter used when the video analyzing means 41 detects a moving object in video data.
  • a moving object detection parameter is used, for example, when the video analyzing means 41 obtains a difference of image data to find an unstill object, and when the video analyzing means 41 uses a filter for removing a moving object other than a person.
  • a moving object detection parameter is a sensitivity threshold, a moving object distinction threshold such as a reciprocating object removal threshold and a shadow removal threshold, and so on.
  • a sensitivity threshold is one of the moving object detection parameters and is a threshold which is the criterion for recognizing a difference between image data of a previous frame and image data of a current frame.
  • a sensitivity threshold is a threshold used for recognizing a moving object (an unstill object).
  • the video analyzing means 41 detects a disparity between image data of a previous frame and image data of a current frame every minimum unit, for example, every pixel.
  • a sensitivity threshold is a threshold for determining the degree of change of the minimum unit for recognizing that (a certain minimum unit of) image data of a current frame has changed from (a certain minimum unit of) image data of a previous frame.
  • a sensitivity threshold is the criterion for determining how to deal with a disparity between image data of a previous frame and image data of a current frame when detecting the disparity (i.e., determining whether or not image data has changed).
  • the video analysis system 1 changes the sensitivity threshold described above based on environment information collected by the environment information collecting means.
  • a sensitivity threshold used when the video analyzing means 41 detects a moving object in video data can be differentiated, for example, between a light place (in the daytime, on a fine day, and so on) and a dark place (at night, on a cloudy day, and so on).
  • a sensitivity threshold can be changed in accordance with environment information such as the time, luminance and weather collected by the environment information collecting means 3 .
  • a sensitivity threshold is set so that change of the minimum unit is recognized at a point when a while color turns to a gray color (i.e., so that a change is detected with a wide disparity range to some degree).
  • time passes by and it gets dark at night in a place where the monitoring camera 2 is installed.
  • the sensitivity threshold for the light place is still used, it may be impossible when, for example, a person wearing navy clothes is moving to detect the person well because the person is under the cover of darkness.
  • the parameter changing means 42 of the video analysis system 1 changes a sensitivity threshold based on the collected environment information. For example, when the environment information collecting means 3 collects environment information of a night time, the parameter changing means 42 changes a sensitivity threshold so that a change of the minimum unit is recognized at a point when a white color turns to a light gray color (i.e., so that a change is detected by a little disparity). Consequently, it becomes possible to accurately detect a small change and, for example, it becomes possible to detect a person wearing navy clothes without any problem even when the person is moving at night.
  • a moving object distinction threshold is one of the moving object detection parameters as well as a sensitivity threshold.
  • a moving object distinction threshold is a threshold which is the criterion used when a filter for removing a moving object other than a detection target object (a person) from a detection target distinguishes a person (a detection target object) from a moving object other than the person (other than the detection target object). Referring to FIG. 5 , it is found that the video analyzing means 41 possibly detects, other than a person as the target of detection, a swinging tree and grass, shadow or the like as moving objects. It is anticipated that detecting all the moving objects and checking what all the detected moving objects are needs execution of a large number of processes and the processing gets heavy.
  • a reciprocating object removal threshold used when a reciprocating object removal filter distinguishes a person from a reciprocating object can be varied, for example, in accordance with environment information such as the presence/absence of the wind and the intensity of the wind collected by the environment information collecting means 3 .
  • environment information such as the presence/absence of the wind and the intensity of the wind collected by the environment information collecting means 3 .
  • the parameter changing means 42 is expected not to set a reciprocating object removal threshold and not to use a reciprocating object removal filter.
  • a tree and grass swing under the influence of the wind.
  • the parameter changing means 42 changes a reciprocating object removal threshold in accordance with the wind speed collected by the environment information collecting means 3 , for example.
  • the parameter changing means 42 is expected to change a reciprocating object removal threshold to be higher.
  • the parameter changing means 42 is expected to set a reciprocating object removal filter to be lower, or expected not to set a reciprocating object removal threshold and not to apply a reciprocating object removal filter.
  • the parameter changing means 42 may be configured to set a reciprocating object removal threshold to be higher in stages in accordance with the intensity of the wind (the wind speed).
  • change of a reciprocating object removal threshold by the parameter changing means 42 in accordance with the environment information (the wind speed) collected by the environment information collecting means 3 makes it possible to accurately remove a reciprocating object from the detection target.
  • a preset threshold (a strong wind determination threshold, a calm determination threshold or the like) may be used for determining whether the environment information of the wind speed collected by the environment information collecting means 3 represents a strong wind or no wind. For example, in a case where the wind speed represented by the environment information is more than the strong wind determination threshold, it is determined windy.
  • the wind speed represented by the environment information is equal to or less than the calm determination threshold, it is determined calm.
  • the intensity of the wind (whether it is windy or calm, for example) based on the environment information of the wind speed.
  • including a plurality of thresholds enables determination of the intensity of the wind in stages based on the environment information of the wind speed.
  • a shadow removal filter is a filter which regards a predetermined shadow as a moving object other than a person to be a detection target and excludes the predetermined shadow from the detection target.
  • a shadow removal threshold is a threshold which is the criterion used when the shadow removal filter distinguishes the person from the predetermined shadow.
  • a shadow removal threshold is, for example, a threshold which represents the amount (length, density or the like) of the shadow to be removed.
  • a shadow removal threshold used when a shadow removal filter distinguishes a person from a shadow can be changed, for example, in accordance with environment information such as time collected by the environment information collecting means 3 .
  • environment information such as time collected by the environment information collecting means 3 .
  • the length of a shadow made around noon is different from the length of a shadow made around dusk. Therefore, the amount such as length of a shadow to be removed is changed in accordance with environment information such as time and latitude collected by the environment information collecting means 3 .
  • the parameter changing means 42 is expected to set a shadow removal threshold to be low.
  • the parameter changing means 42 when time (environment information) collected by the environment information collecting means 3 represents sunrise time or sunset time, the parameter changing means 42 is expected to set a shadow removal threshold to be high.
  • the environment information collecting means 3 may collect the latitude, the longitude, the date or the like, other than time, as environment information.
  • change of a shadow removal threshold by the parameter changing means 42 in accordance with environment information collected by the environment information collecting means 3 makes it possible to accurately remove a shadow.
  • the abovementioned determination using environment information of time is also enabled by using a preset threshold. For example, in a case where environment information of time represents within a range of preset time (a threshold representing a range of time), the parameter changing means is expected to determine that it is sunrise or sunset.
  • a threshold for such a determination may be changed based on environment information of the latitude and longitude, environment information representing season, and so on.
  • a shadow may be different in intensity (density) depending on the weather, for example. Therefore, the parameter changing means 42 may be configured to be capable of changing the amount of a shadow to be removed depending on the weather.
  • the video analyzing means 41 is capable of using various filters for removing a moving object other than a detection target object (a person) from a detection target, and capable of using various moving object distinction thresholds. Meanwhile, the video analyzing means 41 can use various filters other than the ones described above.
  • the video analyzing means 41 may use a color tone correction filter, a sunshine variation filter which is a filter for detecting variation of the whole environment, and so on.
  • the video analyzing means 41 may be configured to also change a moving object distinction threshold of the criterion used when the color tone correction filter or the sunshine variation filter distinguishes a person from a moving object other than the person, based on environment information collected by the environment information collecting means 3 .
  • the video analyzing mean 41 may include a filter for excluding a given value of a white color parameter depending on the intensity of fog, for example.
  • the video analyzing means 41 can be configured to detect a moving object by using a parameter other than the moving object detection parameter described above.
  • the video analyzing means 41 may be configured to regulate a contrast value according to a difference in weather.
  • video data acquired by the monitoring camera 2 includes target regions corresponding to multiple kinds of environment information such as a light portion and a dark portion.
  • the video analyzing means 41 can be configured to separate a sensitivity threshold for the light portion and a sensitivity threshold for the dark portion.
  • the video analyzing means 41 may use one moving object detection parameter for each kind, or may use a plurality of moving object detection parameters for one kind.
  • the video analysis device 4 detects a person of a detection target from the acquired video data and the output device outputs the result of the detection will be described.
  • the monitoring camera 2 acquires video data at a place where the monitoring camera 2 is installed (S 101 ). Then, the monitoring camera 2 having acquired the video data processes the video data into a predetermined format, and thereafter, transmits the video data to the video analyzing means 41 included by the video analysis device 4 via the external network or the like.
  • the video analyzing means 41 of the video analysis device 4 receives the video data transmitted by the monitoring camera 2 . Then, the video analyzing mean 41 detects a person moving in the acquired video data and the trajectory of the moving person by using a moving object detection parameter (S 102 ). Then, the video analyzing means 41 transmits the detection result (the moving person and the trajectory of the moving person) to the analysis result displaying means 43 .
  • the analysis result displaying means 43 receives the detection result transmitted by the video analyzing means 41 . Then, the analysis result displaying means 43 causes the output device such as the monitor connected with the analysis result displaying means 43 to display the analysis result (S 103 ).
  • the video analyzing device 4 detects a person of a detection target from the acquired video data, and the output device outputs the detection result.
  • the video analyzing means 41 detects a moving person and the trajectory of the moving person from video data by using a moving object detection parameter.
  • the video analyzing means 41 receives video data transmitted by the monitoring camera 2 .
  • the video analyzing means 41 then extracts successive image data from the received video data (S 201 ).
  • the video analyzing means 41 obtains, for example, a difference between image data of a previous frame and image data of a current frame and determines whether or not the image data of the current frame has changed from the image data of the previous frame in accordance with a sensitivity threshold (S 202 ).
  • a sensitivity threshold is one of the moving object detection parameters and is the criterion to determine how to deal with a difference between image data of a previous frame and image data of a current frame when the difference is detected (i.e., determine whether there is a change or not). In other words, it varies whether or not the video analyzing means 41 determines that image data of a current frame has changed from image data of a previous frame, in accordance with a sensitivity threshold as one of the moving object detection parameters.
  • the video analyzing means 41 executes processing so as to detect a moving object other than a person as a moving object as little as possible (S 203 ).
  • the video analyzing means 41 distinguishes a person from a moving object other than a person by using a moving object distinction threshold.
  • a moving object distinction threshold is one of the moving object detection parameters and, for example, a reciprocating object removal threshold and a shadow removal threshold fall thereunder.
  • the video analyzing means 41 also checks what the detected object is and detects a person. After that, the video analyzing means 41 acquires the trajectory of movement of the person, and so on.
  • the video analyzing means 41 detects a person of a detection target from video data by using the sensitivity threshold and the moving object distinction threshold of the moving object detection parameters. Then, the video analyzing means 41 transmits the result of the detection to the analysis result displaying means 43 .
  • the environment information collecting means 3 collects environment information representing the external environment of a place where the monitoring camera 2 is installed (S 301 ). Then, the environment information collecting means 3 transmits the collected environment information to the parameter changing means 42 of the video analysis device 4 . Meanwhile, the environment information collecting means 3 may collect environment information by using a physical sensor such as a GPS sensor, or may collect environment information on the Internet or the like.
  • the parameter changing means 42 receives the environment information transmitted by the environment information collecting means 3 . Then, based on the acquired environment information, the parameter changing means 42 changes a moving object detection parameter used when the video analyzing means 41 detects a person from video data (S 302 ). Change of a moving object detection parameter by the parameter changing means 42 is performed, for example, by using a parameter set which is a set of moving object detection parameters previously determined in accordance with a possible external environment.
  • the parameter changing means 42 may be configured to change a moving object detection parameter for each external environment represented by the environment information.
  • the video analysis system 1 includes the environment information collecting means 3 and the parameter changing means 42 .
  • the video analysis system 1 is capable of collecting environment information representing the external environment of a place where the monitoring camera 2 is installed, and changing a moving object detection parameter based on the collected environment information.
  • the video analysis system 1 is capable of changing a moving object detection parameter used by the video analyzing means 41 in accordance with change of the external environment of a place where the monitoring camera 2 is installed.
  • the video analysis system 1 includes a sensitivity threshold as one of the moving object detection parameters. Because the video analysis system 1 thus includes the sensitivity threshold, the environment information collecting means 3 and the parameter changing means 42 , the video analysis system 1 is capable of changing the sensitivity threshold in accordance with the external environment of a place where the monitoring camera 2 is installed. In other words, the video analysis system 1 is capable of changing the criterion for recognizing a disparity of image data when a difference of image data extracted from video data is obtained, in accordance with the external environment. As a result, it is possible to provide the video analysis system 1 that more reliably prevents the accuracy of detection of a moving object from being affected even if the environment of a place where the monitoring camera 2 is installed changes and video data changes.
  • the video analysis system 1 includes a moving object distinction threshold as one of the moving object detection parameters. Because the video analysis system 1 thus includes the moving object distinction threshold, the environment information collecting means 3 and the parameter changing means 42 , the video analysis system 1 is capable of changing a moving object detection threshold in accordance with the external environment of a place where the monitoring camera 2 is installed. In other words, the video analysis system 1 is capable of changing the criterion used when a filter for removing a moving object other than a detection target object from a detection target distinguishes the detection target object from the moving object other than the detection target object, in accordance with the external environment. As a result, it is possible to more reliably provide the video analysis system 1 that prevents the accuracy of detection of a moving object from being affected even if the environment of a place where the monitoring camera 2 is installed changes and video data changes.
  • the video analysis device 4 obtains a difference of image data extracted from video data and thereby detects a person in the video data.
  • the present invention is applicable to a case other than the abovementioned case of detecting a person by obtaining a difference.
  • the present invention is also applicable to a case of detecting a person (a detection target object) by a method such as obtaining a motion vector and using pattern recognition.
  • this exemplary embodiment has described a case where the video analyzing means 41 of the video analysis device 4 detects a person in video data.
  • a detection target object which is a target extracted from video data by the video analyzing means 41 , is not limited to a person.
  • the video analyzing means 41 can detect various targets including a car and an animal such as a dog and a cat as detection target objects.
  • a second exemplary embodiment of the present invention will be described in detail referring to the drawings.
  • a video analysis system 6 including a parameter set which is a set of moving object detection parameters previously determined in accordance with a possible external environment.
  • the video analysis system 6 according to this exemplary embodiment has the partly same configuration as the configuration of the video analysis system 1 described in the first exemplary embodiment. Therefore, in this exemplary embodiment, a description will focus on a different part from the configuration of the first exemplary embodiment.
  • the video analysis system 6 has the monitoring camera 2 , the environment information collecting means 3 , and a video analysis device 5 . Further, the video analysis device 5 has the video analyzing means 41 , the parameter changing means 42 , the analysis result displaying means 43 , and a parameter set storing means 52 . Further, the parameter changing means 42 has an optimum parameter determining means 51 , and a setting updating means 53 . In other words, the video analysis system 6 according to this exemplary embodiment is characterized by having the optimum parameter determining means 51 , the parameter set storing means 52 , and the setting updating means 53 .
  • the same components as those in the first exemplary embodiment will be described, denoted by the same reference numerals as those in the first exemplary embodiment.
  • the optimum parameter determining means 51 has a function of comparing environment information collected by the environment information collecting means 3 with parameter sets stored by the parameter set storing means 52 described later, and determining and selecting the most appropriate parameter set to the acquired condition (the environment information).
  • the optimum parameter determining means 51 receives environment information transmitted by the environment information collecting means 3 . Then, the optimum parameter determining means 51 compares the received environment information with (possible external environments of) parameter sets stored by the parameter set storing means 52 , and determines and selects the most appropriate parameter set to the environment information.
  • Selection of a parameter set by the optimum parameter determining means 51 can be performed by using a table (a matrix) which defines a determination rule, for example.
  • the selection can be performed by storing a combination having ever been experienced by the user as preset values and using the preset values, for example.
  • the optimum parameter determining means 51 can be configured to select a parameter set by using various means. Then, the optimum parameter determining means 51 transmits the parameter set selected by the abovementioned method to the setting updating means 53 .
  • the parameter set storing means 52 is configured by a storage device such as a hard disk and a RAM (Random Access Memory).
  • the parameter set storing means 52 is used for realization of an optimize function previously included by the video analysis system 6 .
  • the parameter set storing means 52 previously stores sets of moving object detection parameters appropriate to possible external environments, as parameter sets.
  • Moving object detection parameters stored as a parameter set are, for example, control of the sensitivity to lightness, a filter for removing the swing of background objects like trees due to the wind, and control of contrast with respect to a difference in weather.
  • the parameter sets stored by the parameter set storing means 52 are compared with the external environment by the optimum parameter determining means 51 , and a parameter set corresponding to the most appropriate possible external environment to the external environment is selected.
  • the parameter set storing means 52 can include a parameter set correcting means for correcting the parameter sets (e.g., values of moving object detection parameters thereof) stored by the parameter set storing means 52 .
  • the setting updating means 53 has a function of setting the most appropriate parameter set to the external environment determined by the optimum parameter determining means 51 dynamically in the video analysis system 6 .
  • the setting updating means 53 receives a parameter set from the optimum parameter determining means 51 .
  • the setting updating means 53 compares current set values (moving object detection parameters being used by the video analyzing means 41 ) with the latest parameter set (received from the optimum parameter determining means 5 ).
  • the setting updating means 53 changes the moving object detection parameters being used by the video analyzing means 41 with the parameter set.
  • the setting updating means 53 does not change the moving object detection parameters but discards the parameter set, for example.
  • the environment information collecting means 3 collects environment information representing the external environment of a place where the monitoring camera 2 is installed (S 401 ). Then, the environment information collecting means 3 transmits the collected environment information to the optimum parameter determining means 51 of the video analysis device 4 .
  • the environment information collecting means 3 may collect environment information by using a physical sensor such as a GPS sensor, or may collect environment information on the Internet or the like.
  • the optimum parameter determining mean 51 receives the environment information transmitted by the environment information collecting means 3 . Then, the optimum parameter determining means 51 compares the external environment represented by the received environment information with possible external environments corresponding to the parameter sets stored by the parameter set storing means 52 (S 402 ). Then, the optimum parameter determining means 51 selects the most appropriate possible external environment to the external environment represented by the environment information (S 403 ), and acquires a parameter set corresponding to the selected possible external environment from the parameter set storing means 52 (S 404 ). After that, the optimum parameter determining means 51 transmits the acquired parameter set to the setting updating means 53 .
  • the setting updating means 53 receives the parameter set transmitted by the optimum parameter determining means 51 . Then, the setting updating means 53 compares moving object detection parameters (current set values) used when the video analyzing means 41 detects a moving object in video data, with the received parameter set (the latest parameter set) (S 405 ). Then, in a case where there is a difference between the current set values and the latest parameter set (YES at S 405 ), the setting updating means 53 changes the moving object detection parameters used by the video analyzing means 41 with the parameter set (S 406 ).
  • the setting updating means 53 does not change the moving object detection parameters but discards the parameter set, for example. In this case, the setting updating means 53 waits until environment information is collected by the environment information collecting means 3 again, and determines whether or not to change moving object detection parameters.
  • the video analysis system 6 includes the optimum parameter determining means 51 , the parameter set storing means 52 , and the setting updating means 53 .
  • the video analysis system 6 is capable of selecting the most appropriate possible external environment to an external environment represented by environment information collected by the environment information collecting means 3 .
  • the video analysis system 6 is capable of changing moving object detection parameters with a parameter set appropriate to the selected possible external environment. As a result, it becomes unnecessary to consider change of each moving object detection parameter in accordance with an external environment represented by the environment information collected by the environment information collecting means 3 , and it is possible to reduce load of changing moving object detection parameters
  • the third exemplary embodiment will describe the overview of the configuration of a video analysis system 7 .
  • the video analysis system 7 has a video data acquiring means 71 , a moving object detecting means 72 , an environment information collecting means 73 , and a parameter changing means 74 .
  • the video data acquiring means 71 has a function of acquiring video data.
  • the video data acquiring means 71 is a monitoring camera or the like and is installed at the airport, at an oil plant, on the street, and so on.
  • the video data acquiring means 71 is connected to the moving object detecting means 72 via a network or the like, and video data acquired by the video data acquiring means 71 is transmitted to the moving object detecting means 72 .
  • the moving object detecting means 72 has a function of detecting, in video data acquired by the video data acquiring means 71 , a moving object moving in the video data by using a moving object detection parameter, which is a parameter for detecting a moving object.
  • the moving object detecting means 72 receives video data transmitted by the video data acquiring means 71 . Subsequently, the moving object detecting means 72 detects a moving object moving in the video data by using a moving object detection parameter. After that, the moving object detecting means 72 outputs the result of the detection to an external device such as a monitor.
  • a moving object detection parameter is a sensitivity threshold, a moving object distinction threshold such as a reciprocating object removal threshold and a shadow removal threshold, or the like. Moreover, a moving object detection parameter is properly changed based on environment information collected by the environment information collecting means 73 as described later.
  • the environment information collecting means 73 has a function of collecting environment information representing the external environment of a place where the video data acquiring means 71 is installed.
  • the environment information collecting means 73 is, for example, a calendar, a Web information collecting means, and various sensors such as a GPS sensor.
  • the environment information collecting means 73 is connected to the parameter changing means 74 , and environment information collected by the environment information collecting means 73 is transmitted to the parameter changing means 74 .
  • the parameter changing means 74 has a function of changing a moving object detection parameter used when the moving object detecting means 72 detects a moving object in video data, based on environment information collected by the environment information collecting means 73 .
  • the parameter changing means 74 receives environment information from the environment information collecting means 73 . Then, the parameter changing means 74 changes a moving object detection parameter based on (an external environment represented by) the acquired environment information (e.g., so as to optimize the moving object detection parameter for the external environment).
  • a moving object detection parameter used when the moving object detecting means 72 detects a moving object in video data is changed based on environment information collected by the environment information collecting means 73 .
  • the video analysis system 7 includes the environment information collecting means 73 and the parameter changing means 74 .
  • the video analysis system 7 is capable of collecting environment information representing the external environment of a place where the video image acquiring means 71 is installed, and changing a moving object detection parameter based on the collected environment information.
  • the video analysis system 7 is capable of changing a moving object detection parameter used by the moving object detecting means 72 in accordance with change of the external environment of a place where the video data acquiring means 71 is installed.
  • the fourth exemplary embodiment describes the configuration overview of a video analysis device 8 .
  • the video analysis device 8 has a video data acquisition part 81 , a moving object detection part 82 , an environment information collection part 83 , and a parameter changing part 84 .
  • the video data acquisition part 81 has a function of acquiring data of a video image shot by a monitoring camera installed outside or the like.
  • the video data acquisition part 81 is connected to a video data acquiring means such as a monitoring camera installed at the airport, at an oil plant, on the street, and so on, via a network or the like, and acquires video data acquired by the video data acquiring means.
  • the video data acquisition part 81 having acquired the video data transmits the acquired video data to the moving object detection part 82 .
  • the moving object detection part 82 has a function of detecting, in video data acquired by the video data acquisition part 81 , a moving object moving in the video data by using a moving object detection parameter, which is a parameter for detecting a moving object.
  • the moving object detection part 82 receives video data transmitted by the video data acquisition part 81 . Subsequently, the moving object detection part 82 detects a moving object moving in the video data by using a moving object detection parameter. After that, the moving object detection part 82 outputs the result of the detection to an external device such as a monitor.
  • a moving object detection parameter is a sensitivity threshold, a moving object distinction threshold such as a reciprocating object removal threshold and a shadow removal threshold, or the like. Moreover, a moving object detection parameter is properly changed based on environment information collected by the environment information collection part 83 as described later.
  • the environment information collection part 83 has a function of acquiring environment information representing the external environment of a place where video data acquired by the video data acquisition part 81 has been acquired.
  • the environment information collection part 83 has, for example, a calendar and a Web information collecting means, and collects environment information from the calendar, the Web, and so on.
  • the environment information collection part 83 is connected with various sensors such as a GPS sensor which acquires environment information such as meteorological information (time, latitude and longitude, weather, wind speed, and so on) of the place where the video data has been acquired.
  • the environment information collection part 83 collects environment information from the various sensors. Then, the environment information collecting part 83 having collected environment information transmits the collected environment information to the parameter changing part 84 .
  • the parameter changing part 84 has a function of changing a moving object detection parameter used when the moving object detection part 82 detects a moving object in video data, based on environment information collected by the environment information collection part 83 .
  • the parameter changing part 84 receives environment information from the environment information collection part 83 .
  • the parameter changing part 84 changes a moving object detection parameter based on (an external environment represented by) the acquired environment information (e.g., so as to optimize a moving object detection parameter for the external environment). Consequently, a moving object detection parameter used when the moving object detection part 82 detects a moving object in video data is changed based on environment information collected by the environment information collection part 83 .
  • the video analysis device 8 includes the environment information collection part 83 and the parameter changing part 84 .
  • the video analysis system 8 is capable of collecting environment information representing the external environment of a place where video data acquired by of the video image acquisition part 81 has been acquired, and changing a moving object detection parameter based on the collected environment information.
  • the video analysis device 8 is capable of changing a moving object detection parameter used by the moving object detection part 82 in accordance with change of the external environment of a place where video data acquired by the video data acquisition part 81 has been acquired.
  • a program as another aspect of the present invention is a computer program for causing a video analysis device to function as: a video data acquisition part which acquires video data acquired by an external device from the external device; a moving object detection part which detects, in video data acquired by the video data acquisition part, a moving object moving in the video data by using a moving object detection parameter, which is a parameter for detecting a moving object; an environment information collection part which collects environment information representing the external environment of a place where the external device having acquired the video data; and a parameter changing part which changes a moving object detection parameter used when the moving object detection part detects a moving object, based on the environment information acquired by the environment information collection part.
  • a method for video analysis executed by operation of the abovementioned video analysis device 8 is a method including: collecting environment information representing the external environment of a place where an external device acquiring video data is installed; changing a moving object detection parameter used when detecting a moving object in video data, based on the collected environment information; and detecting, in video data acquired by an external device, the moving object moving in the video data by using the changed moving object detection parameter.
  • the invention of a program or a video analysis method having the abovementioned configurations has the same actions as the video analysis device 8 , and therefore, can achieve the object of the present invention described above.
  • a video analysis system comprising:
  • a moving object detecting means for detecting a moving object from video data by using a moving object detection parameter, the video data having been acquired by the video data acquiring means, the moving object detection parameter being a parameter for detecting a moving object;
  • an environment information collecting means for collecting environment information representing an external environment of a place where the video data acquiring means is installed
  • a parameter changing means for changing the moving object detection parameter on a basis of the environment information collected by the environment information collecting means, the moving object detection parameter being used when the moving object detecting means detects the moving object.
  • the video analysis system includes the environment information collecting means and the parameter changing means.
  • Such a configuration enables the video analysis system to collect environment information representing the external environment of a place where the video data acquiring means is installed and change the moving object detection parameter based on the collected environment information.
  • such a configuration enables the video analysis system to change the moving object detection parameter used by the moving object detecting means in accordance with change of the external environment of the place where the video data acquiring means is installed.
  • the moving object detecting means is configured to detect the moving object by obtaining a difference of image data extracted from the video data and includes a sensitivity threshold as one of moving object detection parameters, the sensitivity threshold being a predetermined threshold to become a criterion for recognizing a disparity between image data of a previous frame and image data of a current frame; and
  • the parameter changing means changes the sensitivity threshold on a basis of the environment information collected by the environment information collecting means, the sensitivity threshold being one of the moving object detection parameters.
  • the video analysis system includes the sensitivity threshold as one of the moving object detection parameters.
  • the video analysis system enables the video analysis system to change the sensitivity threshold in accordance with the external environment of a place where the video data acquiring means is installed.
  • the video analysis system enables the video analysis system to change the criterion for recognizing a disparity of image data when obtaining a difference of image data extracted from video data, in accordance with the external environment.
  • the moving object detecting means is configured to use a predetermined filter when detecting the moving object from the video data and includes a moving object distinction threshold as one of moving object detection parameters, the predetermined filter being for removing a moving object other than a detection target object from a detection target, the detection target object being a target to be detected, the moving object distinction threshold being a predetermined threshold used when the predetermined filter distinguishes the detection target object from the moving object other than the detection target object; and
  • the parameter changing means changes the moving object distinction threshold on a basis of the environment information collected by the environment information collecting means, the moving object distinction threshold being one of the moving object detection parameters.
  • the video analysis system includes the moving object distinction threshold as one of the moving object detection parameters.
  • Such a configuration enables the video analysis system 1 to change the moving object detection threshold in accordance with the external environment of a place where the monitoring camera 2 is installed.
  • such a configuration enables the video analysis system 1 to change the criterion used when the filter for removing a moving object other than a detection target object from the target of detection distinguishes a detection target object from a moving object other than the detection target object, in accordance with the external environment.
  • it is possible to provide the video analysis system 1 which more reliably prevents the accuracy of detection of a moving object from being affected even when the environment of the place where the monitoring camera is installed changes and video data changes.
  • the moving object detecting means is configured to use a reciprocating object removal filter when detecting the moving object from the video data and includes a reciprocating object removal threshold as one of the moving object distinction parameters, the reciprocating object removal filter being configured to regard a reciprocating object as the moving object other than the detection target object and remove the reciprocating object from the detection target, the reciprocating object performing predetermined reciprocation, the predetermined reciprocation being a repetitive movement, the reciprocating object removal threshold being a threshold used when the reciprocating object removal filter distinguishes the detection target object from the reciprocating object; and
  • the parameter changing means changes the reciprocating object removal threshold on a basis of the environment information collected by the environment information collecting means, the reciprocating object removal threshold being one of the moving object detection parameters.
  • the video analysis system includes the reciprocating object removal threshold as one of the moving object distinction thresholds.
  • Such a configuration enables the video analysis system to change the reciprocating object removal threshold in accordance with the external environment of a place where the video data acquiring means is installed.
  • the video analysis system enables the video analysis system to change the criterion used when the filter for removing a moving object other than a detection target object, that is, removing a reciprocating object from the target of detection distinguishes a detection target object from a moving object other than the detection target object, that is, from a reciprocating object, in accordance with the external environment.
  • it is possible to provide the video analysis system which more reliably prevents the accuracy of detection of a moving object from being affected even when the environment of the place where the video data acquiring means is installed changes and video data changes.
  • the environment information collecting means is configured to collect wind speed information as the environment information, the wind speed information representing a wind speed;
  • the parameter changing means changes the reciprocating object removal threshold in accordance with the wind speed represented by the wind speed information collected by the environment information collecting means, the reciprocating object removal threshold being one of the moving object detection parameters.
  • the moving object detecting means is configured to use a shadow removal filter when detecting the moving object from the video data and includes a shadow removal threshold as one of moving object distinction parameters, the shadow removal filter being configured to regard a predetermined shadow as the moving object other than the detection target object and remove the predetermined shadow from the detection target, the shadow removal threshold being a threshold used when the shadow removal filter distinguishes the detection target object from the predetermined shadow; and
  • the parameter changing means changes the shadow removal threshold on a basis of the environment information collected by the environment information collecting means, the shadow removal threshold being one of the moving object detection parameters.
  • the video analysis system includes the shadow removal threshold as one of the moving object distinction thresholds.
  • Such a configuration enables the video analysis system to change the shadow removal threshold in accordance with the external environment of a place where the video data acquiring means is installed.
  • such a configuration enables the video analysis system to change the criterion used when the filter for removing a moving object other than a detection target object, that is, removing a shadow from the target of detection distinguishes a detection target object from a moving object other than the detection target object, that is, from a shadow, in accordance with the external environment.
  • the environment information collecting means is configured to collect time information as the environment information, the time information representing time;
  • the parameter changing means changes the shadow removal threshold in accordance with the time represented by the time information collected by the environment information collecting means, the shadow removal threshold being one of the moving object detection parameters.
  • the video analysis system further comprising a parameter set storing means for storing a predetermined set of the moving object detection parameters as a parameter set, the predetermined set of the moving object detection parameters being corresponding to a possible external environment,
  • the parameter changing means changes the moving object detection parameters, on a basis of the environment information collected by the environment information collecting means, by using the parameter set corresponding to the external environment represented by the environment information.
  • the video analysis system includes the parameter set storing means, and is configured to change the moving object detection parameters by using a parameter set corresponding to an external environment represented by environment information.
  • a configuration enables the video analysis system to select a possible external environment appropriate to the external environment represented by the environment information.
  • such a configuration enables the video analysis system to change the moving object detection parameters by using a parameter set corresponding to the selected possible external environment. As a result, it is not necessary to consider change of the individual moving object detection parameters in accordance with the external environment represented by the environment information, and it is possible to achieve reduction of the load for changing the moving object detection parameters.
  • the video analysis system according to any of Supplementary Notes 1 to 8, wherein the environment information is at least one of: luminance; time and date; meteorological information; and latitude and longitude.
  • a video analysis device comprising:
  • a video data acquisition part configured to acquire video data from an external device, the video data having been acquired by the external device
  • a moving object detection part configured to detect a moving object from video data by using a moving object detection parameter, the video data having been acquired by the video data acquisition part, the moving object detection parameter being a parameter for detecting a moving object;
  • an environment information collection part configured to collect environment information representing an external environment of a place where the external device having acquired the video data is installed
  • a parameter changing part configured to change the moving object detection parameter on a basis of the environment information collected by the environment information collection part, the moving object detection parameter being used when the moving object detection part detects the moving object.
  • the moving object detection part is configured to detect the moving object by obtaining a difference of image data extracted from the video data and includes a sensitivity threshold as one of moving object detection parameters, the sensitivity threshold being a predetermined threshold to become a criterion for recognizing a disparity between image data of a previous frame and image data of a current frame; and
  • the parameter changing part changes the sensitivity threshold on a basis of the environment information of the place where the external device having acquired the video data is installed, the sensitivity threshold being one of the moving object detection parameters.
  • the moving object detection part is configured to use a predetermined filter when detecting the moving object from the video data and includes a moving object distinction threshold as one of moving object detection parameters, the predetermined filter being for removing a moving object other than a detection target object from a detection target, the detection target object being a target of detection, the moving object distinction threshold being a predetermined threshold used when the predetermined filter distinguishes the detection target object from the moving object other than the detection target object; and
  • the parameter changing part changes the moving object distinction threshold on a basis of the environment information of the place where the external device having acquired the video data is installed, the moving object distinction threshold being one of the moving object detection parameters.
  • the moving object detection part is configured to use a reciprocating object removal filter when detecting the moving object from the video data and includes a reciprocating object removal threshold as one of the moving object distinction parameters, the reciprocating object removal filter being configured to regard a reciprocating object as the moving object other than the detection target object and remove the reciprocating object from the detection target, the reciprocating object performing predetermined reciprocation, the predetermined reciprocation being a repetitive movement, the reciprocating object removal threshold being a threshold used when the reciprocating object removal filter distinguishes the detection target object from the reciprocating object; and
  • the parameter changing part changes the reciprocating object removal threshold on a basis of the environment information of the place where the external device having acquired the video data is installed, the reciprocating object removal threshold being one of the moving object detection parameters.
  • the moving object detection part is configured to use a shadow removal filter when detecting the moving object from the video data and includes a shadow removal threshold as one of the moving object distinction parameters, the shadow removal filter being configured to regard a predetermined shadow as the moving object other than the detection target object and remove the predetermined shadow from the detection target, the shadow removal threshold being a threshold used when the shadow removal filter distinguishes the detection target object from the predetermined shadow; and
  • the parameter changing part changes the shadow removal threshold on a basis of the environment information of the place where the external device having acquired the video data is installed, the shadow removal threshold being one of the moving object detection parameters.
  • the video analysis device further comprising a parameter set storage part configured to store a predetermined set of the moving object detection parameters as a parameter set, the predetermined set of the moving object detection parameters being corresponding to a possible external environment,
  • the parameter changing part changes the moving object detection parameters, on a basis of the environment information of the place where the external device having acquired the video data is installed, by using the parameter set corresponding to an external environment represented by the environment information.
  • the video analysis device according to any of Supplementary Notes 10 to 15, wherein the environment information is at least one of: luminance; time and date; meteorological information; and latitude and longitude.
  • a video analysis method comprising:
  • the environment information representing an external environment of a place where an external device acquiring video data is installed, the moving object detection parameter being used when detecting a moving object from the video data;
  • the sensitivity threshold being used as one of moving object detection parameters, the sensitivity threshold being a predetermined threshold to become a criterion for recognizing a disparity between image data of a previous frame and image data of a current frame;
  • the moving object distinction threshold being used as one of the moving object detection parameters, the moving object distinction threshold being a predetermined threshold used when distinguishing a detection target object to become a target of detection from a moving object other than the detection target object;
  • the video analysis method according to any of Supplementary Notes 17 to 19, wherein the environment information is at least one of: luminance; time and date; meteorological information; and latitude and longitude.
  • a computer program including instructions for causing a video analysis device to function as:
  • a video data acquisition part configured to acquire video data from an external device, the video data being acquired by the external device
  • a moving object detection part configured to detect a moving object from video data by using a moving object detection parameter, the moving object moving in the video data, the video data being acquired by the video data acquisition part, the moving object detection parameter being a parameter for detecting a moving object;
  • an environment information collection part configured to collect environment information representing an external environment of a place where the external device having acquired the video data is installed
  • a parameter changing part configured to change the moving object detection parameter on a basis of the environment information acquired by the environment information collection part, the moving object detection parameter being used when the moving object detection part detects the moving object.
  • the program described in the exemplary embodiments and the supplementary note is stored in the storage device or recorded on a computer-readable medium.
  • the recording medium is a portable medium such as a flexible disk, an optical disk, a magnet-optical disk and a semiconductor memory.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Studio Devices (AREA)
US15/117,814 2014-02-14 2015-02-05 Video analysis system Abandoned US20170011529A1 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
JP2014026465 2014-02-14
JP2014-026465 2014-02-14
PCT/JP2015/000529 WO2015122161A1 (ja) 2014-02-14 2015-02-05 映像解析システム

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2015/000529 A-371-Of-International WO2015122161A1 (ja) 2014-02-14 2015-02-05 映像解析システム

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US16/354,347 Continuation US10636129B2 (en) 2014-02-14 2019-03-15 Wind motion threshold image analysis system
US16/354,372 Continuation US10706515B2 (en) 2014-02-14 2019-03-15 Shadow removal video analysis system

Publications (1)

Publication Number Publication Date
US20170011529A1 true US20170011529A1 (en) 2017-01-12

Family

ID=53799912

Family Applications (4)

Application Number Title Priority Date Filing Date
US15/117,814 Abandoned US20170011529A1 (en) 2014-02-14 2015-02-05 Video analysis system
US16/354,372 Active US10706515B2 (en) 2014-02-14 2019-03-15 Shadow removal video analysis system
US16/354,347 Active US10636129B2 (en) 2014-02-14 2019-03-15 Wind motion threshold image analysis system
US16/891,555 Active 2035-03-07 US11398018B2 (en) 2014-02-14 2020-06-03 Video shadow and motion removal system

Family Applications After (3)

Application Number Title Priority Date Filing Date
US16/354,372 Active US10706515B2 (en) 2014-02-14 2019-03-15 Shadow removal video analysis system
US16/354,347 Active US10636129B2 (en) 2014-02-14 2019-03-15 Wind motion threshold image analysis system
US16/891,555 Active 2035-03-07 US11398018B2 (en) 2014-02-14 2020-06-03 Video shadow and motion removal system

Country Status (5)

Country Link
US (4) US20170011529A1 (ja)
JP (1) JP6350549B2 (ja)
AR (1) AR099429A1 (ja)
SG (1) SG11201606716VA (ja)
WO (1) WO2015122161A1 (ja)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150279009A1 (en) * 2014-03-31 2015-10-01 Sony Corporation Image processing apparatus, image processing method, and program
US20170013230A1 (en) * 2014-02-14 2017-01-12 Nec Corporation Video processing system
JP2017169076A (ja) * 2016-03-17 2017-09-21 セコム株式会社 飛行物体監視システム
US20180276485A1 (en) * 2016-09-14 2018-09-27 Nauto Global Limited Systems and methods for safe route determination
US10268909B2 (en) * 2016-09-14 2019-04-23 Nauto, Inc. Systems and methods for near-crash determination
US10503990B2 (en) 2016-07-05 2019-12-10 Nauto, Inc. System and method for determining probability that a vehicle driver is associated with a driver identifier
US20200105042A1 (en) * 2018-09-28 2020-04-02 Mana AKAIKE Display control system, display control device, and display control method
US10703268B2 (en) 2016-11-07 2020-07-07 Nauto, Inc. System and method for driver distraction determination
CN111417993A (zh) * 2017-11-13 2020-07-14 罗伯特·博世有限公司 用于提供至少一个对象的位置的方法和设备
US10956752B2 (en) * 2016-11-30 2021-03-23 Robert Bosch Gmbh Camera for monitoring a monitored area and monitoring device, and method for monitoring a monitored area
US11017479B2 (en) 2017-06-16 2021-05-25 Nauto, Inc. System and method for adverse vehicle event determination
US11175145B2 (en) 2016-08-09 2021-11-16 Nauto, Inc. System and method for precision localization and mapping
US11363201B2 (en) 2020-04-28 2022-06-14 Axis Ab Statistics-based electronic image stabilization
US11392131B2 (en) 2018-02-27 2022-07-19 Nauto, Inc. Method for determining driving policy

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11758238B2 (en) * 2014-12-13 2023-09-12 Fox Sports Productions, Llc Systems and methods for displaying wind characteristics and effects within a broadcast
US10550252B2 (en) * 2017-04-20 2020-02-04 Chevron Phillips Chemical Company Lp Bimodal PE resins with improved melt strength
KR102402883B1 (ko) * 2017-06-20 2022-05-27 한화테크윈 주식회사 객체 검출 시스템, 방법 및 컴퓨터 판독 가능한 기록매체
JP7054771B2 (ja) * 2017-08-29 2022-04-15 パナソニックＩｐマネジメント株式会社 物体検出システム、プログラム及び物体検出方法
WO2019045091A1 (ja) * 2017-09-04 2019-03-07 日本電気株式会社 情報処理装置、計数システム、計数方法およびプログラム記憶媒体
JP6934386B2 (ja) * 2017-10-03 2021-09-15 日本放送協会 動体追尾装置及びそのプログラム
JP7338174B2 (ja) * 2019-03-06 2023-09-05 オムロン株式会社 物体検出装置および物体検出方法
JP6742623B1 (ja) * 2019-11-13 2020-08-19 尚範 伊達 監視装置、監視方法、及びプログラム
US11482049B1 (en) 2020-04-14 2022-10-25 Bank Of America Corporation Media verification system
US11308646B2 (en) * 2020-05-13 2022-04-19 Walmart Apollo, Llc Extracting color from item images
US11594032B1 (en) 2021-02-17 2023-02-28 Bank Of America Corporation Media player and video verification system
US11527106B1 (en) 2021-02-17 2022-12-13 Bank Of America Corporation Automated video verification
US11790694B1 (en) 2021-02-17 2023-10-17 Bank Of America Corporation Video player for secured video stream
US11928187B1 (en) 2021-02-17 2024-03-12 Bank Of America Corporation Media hosting system employing a secured video stream
US11941051B1 (en) 2021-06-24 2024-03-26 Bank Of America Corporation System for performing programmatic operations using an image-based query language
US11526548B1 (en) 2021-06-24 2022-12-13 Bank Of America Corporation Image-based query language system for performing database operations on images and videos
US11784975B1 (en) 2021-07-06 2023-10-10 Bank Of America Corporation Image-based firewall system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020071034A1 (en) * 2000-10-31 2002-06-13 Wataru Ito Intruding object detection method and intruding object monitor apparatus which automatically set a threshold for object detection
US20100063672A1 (en) * 2008-09-11 2010-03-11 Noel Wayne Anderson Vehicle with high integrity perception system
US20100231721A1 (en) * 2007-11-30 2010-09-16 Searidge Technologies Inc. Airport target tracking system
US20100290710A1 (en) * 2009-04-22 2010-11-18 Nikhil Gagvani System and method for motion detection in a surveillance video
US20120106797A1 (en) * 2010-08-03 2012-05-03 Empire Technology Development Llc Identification of objects in a video
US20130136350A1 (en) * 2011-11-28 2013-05-30 Industrial Technology Research Institute Apparatus and method for foreground detection
US20140270548A1 (en) * 2013-03-13 2014-09-18 Denso Corporation Object detection apparatus

Family Cites Families (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04358291A (ja) * 1991-06-04 1992-12-11 Hitachi Ltd 色彩画像変更方法
JP4343325B2 (ja) * 1999-05-14 2009-10-14 富士通株式会社 移動物体検知装置
US6754367B1 (en) 1999-09-30 2004-06-22 Hitachi Denshi Kabushiki Kaisha Method and apparatus for automatically detecting intrusion object into view of image pickup device
JP2001160146A (ja) * 1999-12-01 2001-06-12 Matsushita Electric Ind Co Ltd 画像認識方法および画像認識装置
JP2002074370A (ja) * 2000-08-28 2002-03-15 Ntt Data Corp 動画像による監視システム、方法及びコンピュータ読み取り可能な記録媒体
JP3961344B2 (ja) * 2002-05-31 2007-08-22 株式会社東芝 画像処理システム、画像処理方法、画像処理プログラム
JP4296287B2 (ja) * 2003-05-27 2009-07-15 日産自動車株式会社 車両認識装置
US7574043B2 (en) 2005-06-27 2009-08-11 Mitsubishi Electric Research Laboratories, Inc. Method for modeling cast shadows in videos
US7639878B2 (en) * 2005-11-17 2009-12-29 Honeywell International Inc. Shadow detection in images
US8045761B2 (en) * 2006-05-30 2011-10-25 Intelliview Technologies Inc. Detection of environmental conditions in a sequence of images
US8280106B2 (en) 2007-09-29 2012-10-02 Samsung Electronics Co., Ltd. Shadow and highlight detection system and method of the same in surveillance camera and recording medium thereof
US20100111374A1 (en) * 2008-08-06 2010-05-06 Adrian Stoica Method for using information in human shadows and their dynamics
TWI368185B (en) 2008-11-06 2012-07-11 Ind Tech Res Inst Method for detecting shadow of object
IL196161A (en) * 2008-12-24 2015-03-31 Rafael Advanced Defense Sys Download shadows from images within a video signal
JP5196178B2 (ja) 2009-01-21 2013-05-15 日本電気株式会社 画像補正システム、画像補正サーバ、画像補正方法、画像補正プログラム
JP5360403B2 (ja) 2009-09-29 2013-12-04 サクサ株式会社 移動体撮像装置
JP5663946B2 (ja) 2010-05-13 2015-02-04 トヨタ自動車株式会社 車両走行制御装置
JP5647459B2 (ja) * 2010-08-13 2014-12-24 セコム株式会社 移動物体検出装置
US20130128050A1 (en) * 2011-11-22 2013-05-23 Farzin Aghdasi Geographic map based control
WO2013096493A1 (en) * 2011-12-19 2013-06-27 Younis Saed G Remotely sensing and adapting irrigation system
US8744125B2 (en) * 2011-12-28 2014-06-03 Pelco, Inc. Clustering-based object classification
US9681125B2 (en) 2011-12-29 2017-06-13 Pelco, Inc Method and system for video coding with noise filtering
US9530221B2 (en) 2012-01-06 2016-12-27 Pelco, Inc. Context aware moving object detection
KR101735565B1 (ko) 2012-06-25 2017-05-15 한화테크윈 주식회사 열 그림자를 제거할 수 있는 움직임 검출 시스템 및 방법

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020071034A1 (en) * 2000-10-31 2002-06-13 Wataru Ito Intruding object detection method and intruding object monitor apparatus which automatically set a threshold for object detection
US20100231721A1 (en) * 2007-11-30 2010-09-16 Searidge Technologies Inc. Airport target tracking system
US20100063672A1 (en) * 2008-09-11 2010-03-11 Noel Wayne Anderson Vehicle with high integrity perception system
US20100290710A1 (en) * 2009-04-22 2010-11-18 Nikhil Gagvani System and method for motion detection in a surveillance video
US20120106797A1 (en) * 2010-08-03 2012-05-03 Empire Technology Development Llc Identification of objects in a video
US20130136350A1 (en) * 2011-11-28 2013-05-30 Industrial Technology Research Institute Apparatus and method for foreground detection
US20140270548A1 (en) * 2013-03-13 2014-09-18 Denso Corporation Object detection apparatus

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170013230A1 (en) * 2014-02-14 2017-01-12 Nec Corporation Video processing system
US11665311B2 (en) 2014-02-14 2023-05-30 Nec Corporation Video processing system
US10389969B2 (en) * 2014-02-14 2019-08-20 Nec Corporation Video processing system
US20150279009A1 (en) * 2014-03-31 2015-10-01 Sony Corporation Image processing apparatus, image processing method, and program
JP2017169076A (ja) * 2016-03-17 2017-09-21 セコム株式会社 飛行物体監視システム
US11580756B2 (en) 2016-07-05 2023-02-14 Nauto, Inc. System and method for determining probability that a vehicle driver is associated with a driver identifier
US10503990B2 (en) 2016-07-05 2019-12-10 Nauto, Inc. System and method for determining probability that a vehicle driver is associated with a driver identifier
US11175145B2 (en) 2016-08-09 2021-11-16 Nauto, Inc. System and method for precision localization and mapping
US10733460B2 (en) * 2016-09-14 2020-08-04 Nauto, Inc. Systems and methods for safe route determination
US11023751B2 (en) * 2016-09-14 2021-06-01 Nauto, Inc. Systems and methods for safe route determination
US10769456B2 (en) * 2016-09-14 2020-09-08 Nauto, Inc. Systems and methods for near-crash determination
US20180276485A1 (en) * 2016-09-14 2018-09-27 Nauto Global Limited Systems and methods for safe route determination
US10268909B2 (en) * 2016-09-14 2019-04-23 Nauto, Inc. Systems and methods for near-crash determination
US10703268B2 (en) 2016-11-07 2020-07-07 Nauto, Inc. System and method for driver distraction determination
US11485284B2 (en) 2016-11-07 2022-11-01 Nauto, Inc. System and method for driver distraction determination
US10956752B2 (en) * 2016-11-30 2021-03-23 Robert Bosch Gmbh Camera for monitoring a monitored area and monitoring device, and method for monitoring a monitored area
US11164259B2 (en) 2017-06-16 2021-11-02 Nauto, Inc. System and method for adverse vehicle event determination
US11017479B2 (en) 2017-06-16 2021-05-25 Nauto, Inc. System and method for adverse vehicle event determination
CN111417993A (zh) * 2017-11-13 2020-07-14 罗伯特·博世有限公司 用于提供至少一个对象的位置的方法和设备
US11392131B2 (en) 2018-02-27 2022-07-19 Nauto, Inc. Method for determining driving policy
US20200105042A1 (en) * 2018-09-28 2020-04-02 Mana AKAIKE Display control system, display control device, and display control method
US11017578B2 (en) * 2018-09-28 2021-05-25 Ricoh Company, Ltd. Display control system to control a display based on detecting wind
US11363201B2 (en) 2020-04-28 2022-06-14 Axis Ab Statistics-based electronic image stabilization

Also Published As

Publication number Publication date
BR112016018554A2 (ja) 2018-05-02
US20190213721A1 (en) 2019-07-11
US20200294212A1 (en) 2020-09-17
SG11201606716VA (en) 2016-09-29
US11398018B2 (en) 2022-07-26
JPWO2015122161A1 (ja) 2017-03-30
AR099429A1 (es) 2016-07-20
US10636129B2 (en) 2020-04-28
JP6350549B2 (ja) 2018-07-04
WO2015122161A1 (ja) 2015-08-20
US20190213720A1 (en) 2019-07-11
US10706515B2 (en) 2020-07-07

Similar Documents

Publication Publication Date Title
US11398018B2 (en) Video shadow and motion removal system
US10636173B1 (en) Dynamic calibration of surveillance devices
KR102478335B1 (ko) 채널별 객체 검출 최적화를 위한 영상분석 방법 및 서버장치
US11386536B2 (en) Image analysis device, image analysis method and image analysis program
US11012683B1 (en) Dynamic calibration of surveillance devices
US8411947B2 (en) Video processing to detect movement of an object in the scene
EP3582181B1 (en) Method, device and system for determining whether pixel positions in an image frame belong to a background or a foreground
KR101339026B1 (ko) 열화상 카메라 가시성 개선 방법 및 장치
KR102391853B1 (ko) 영상 정보 처리 시스템 및 방법
CN111444758A (zh) 一种基于时空信息的行人重识别方法及装置
CN113935395A (zh) 对象识别神经网络的训练
CN111428626B (zh) 一种移动物体的识别方法、装置、及存储介质
KR102171384B1 (ko) 영상 보정 필터를 이용한 객체 인식 시스템 및 방법
JP2021007055A (ja) 識別器学習装置、識別器学習方法およびコンピュータプログラム
KR102457470B1 (ko) 영상 분석을 이용한 인공지능 기반 강수 판단장치 및 방법
CN114040124B (zh) 森林草原巡视无人机的相机自适应曝光处理系统及方法
JP7130488B2 (ja) 画像処理装置、画像処理方法、および、画像処理プログラム
JP5947588B2 (ja) 監視装置
CN111582025B (zh) 一种移动物体的识别方法、装置、及存储介质
CN113963502B (zh) 一种全天候违法行为自动化巡检方法及系统
EP4276770A1 (en) Object re-identification in video streams
JP5804155B1 (ja) 情報処理装置、情報処理方法およびプログラム
BR112016018554B1 (pt) Dispositivo de análise de vídeo e processo de análise de vídeo
JP2007158607A (ja) 監視装置
JP2019052904A (ja) ランドマーク検出方法及びランドマーク検出装置

Legal Events

Date Code Title Description
AS Assignment

Owner name: NEC CORPORATION, JAPAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:URASHITA, KEIICHI;REEL/FRAME:039393/0155

Effective date: 20160801

STPP Information on status: patent application and granting procedure in general

Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION

STPP Information on status: patent application and granting procedure in general

Free format text: NON FINAL ACTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER

STPP Information on status: patent application and granting procedure in general

Free format text: FINAL REJECTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: ADVISORY ACTION MAILED

STCB Information on status: application discontinuation

Free format text: EXPRESSLY ABANDONED -- DURING EXAMINATION