Image data processing apparatus and method for object detection and judging suspicious objects

• US 8,218,814 B2
• Filed: 03/28/2008
• Issued: 07/10/2012
• Est. Priority Date: 03/30/2007
• Status: Active Grant

• Current Assignee: Hitachi Kokusai Electric Incorporated (Hitachi, Ltd.)
• Original Assignee: Hitachi Kokusai Electric Incorporated (Hitachi, Ltd.)
• Inventors: Ito, Mitsue, Ito, Wataru, Takahashi, Kazuya
• Primary Examiner(s): Repko, Jason M
• Assistant Examiner(s): Thomas, Mia M
• Application Number:
  US12/057,823
• Publication Number:
  US 20080247601A1
• Time in Patent Office: 1,565 Days
• Field of Search: 382/100, 382/103, 382/190, 382/286, 382/289, 348/135, 348/142, 348/143, 348/152, 340/540
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 2207/30241 Trajectory
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction

Abstract

An image processing apparatus such as a monitor system for executing image processing to present a suspicious object effectively. An object detecting unit detects an object contained in an image, an associating unit associates a plurality of objects detected with the object detecting unit, with each other, and an evaluating unit evaluates (e.g., evaluation as being suspicious) an object detected by the object detecting unit, and an association evaluating unit evaluates another object associated by the associating unit with the object evaluated by the evaluating unit, in accordance with the evaluation made by the evaluating unit.

Claims

1. An image processing apparatus comprising:

object detecting means for detecting each object of a plurality of objects contained in a plurality of images arrayed in order of points in time at which said images are taken, respectively;

tracking means for tracking an object by connecting two objects which are respectively detected from two images taken at different points in time among said plurality of images and which can be considered to be a same object;

associating means for associating said plurality of objects with each other based on a result of tracking objects by said tracking means;

evaluating means for giving evaluations to said each object detected with said object detecting means, respectively; and

association evaluating means for giving an evaluation to another object which has been associated by said associating means with the object to which the evaluation has been given by said evaluating means, based on the evaluations made by said evaluating means,

wherein said associating means executes association when one or more of the following conditions occur including a condition of associating objects in a case in which said objects are coupled into one object, a condition of associating objects in a case in which one object is separated into said objects, and a condition of associating objects in a case in which distances among said objects become shorter than a predetermined threshold value.

2. The image processing apparatus according to claim 1, further comprising:

displaying means for displaying said each object detected with said object detecting means in a display style representative of association of said plurality of objects.

3. The image processing apparatus according to claim 1, wherein said predetermined judgment criterion includes whether each of said objects exists in a specific area of the image.

4. The image processing apparatus according to claim 1, wherein the evaluations made by the evaluating means include a level of danger or suspicion, and

wherein the association evaluating means is configured to give an evaluation to another object (A) which has been associated with the object (B) to which the evaluation has been given by said evaluating means, said evaluation given to said another object (A) having a lowered level of danger or suspicion than said evaluation given to said object (B).

5. The image processing apparatus according to claim 4, further comprising:

image recording means for storing said plurality of images and detected-object information including respective times of detection of the detected objects and evaluations for the detected objects;

display processing means for displaying said detected objects by drawing said detected objects in a display style corresponding to the evaluations given to said detected objects,

wherein said object detecting means is configured to detect, as said objects, clusters of pixels detected from binary images each obtained by processing said plurality of images,

wherein said evaluating means is configured to either give a predetermined evaluation to the detected object which entered a predetermined area, give a predetermined evaluation to the detected object which has a predetermined motion speed or moving direction, or give a predetermined evaluation to the detected object which stayed in a predetermined area for a predetermined period of time, and

wherein said associating means is configured to store the associations of said plurality of objects into said image recording means as a table or a search tree.

6. The image processing apparatus according to claim 4, wherein one detected object is separated into a plurality of detected objects can be detected, if a plurality of detected objects are detected in some frame at the positions whose distances from a position of one detected object in the previous frame are not larger than, or smaller than, a predetermined threshold value, and if the plurality of detected objects are not collected from the surrounding.

7. The image processing apparatus according to claim 1, further comprising:

image recording means for storing said plurality of images and detected-object information including respective times of detection of the detected objects and evaluations for the detected objects;

display processing means for displaying said detected objects by drawing said detected objects in a display style corresponding to the evaluations given to said detected objects,

wherein said object detecting means is configured to detect, as said objects, clusters of pixels detected from binary images each obtained by processing said plurality of images,

wherein said evaluating means is configured to either give a predetermined evaluation to the detected object which entered a predetermined area, give a predetermined evaluation to the detected object which has a predetermined motion speed or moving direction, or give a predetermined evaluation to the detected object which stayed in a predetermined area for a predetermined period of time, and

wherein said associating means is configured to store the associations of said plurality of objects into said image recording means as a table or a search tree.

8. The image processing apparatus according to claim 1, wherein one detected object is separated into a plurality of detected objects can be detected, if a plurality of detected objects are detected in some frame at the positions whose distances from a position of one detected object in the previous frame are not larger than, or smaller than, a predetermined threshold value, and if the plurality of detected objects are not collected from the surrounding.

9. An image processing method comprising the steps of:

detecting each object of a plurality of objects contained in a plurality of images arrayed in order of points in time at which said images are taken, respectively;

tracking an object by connecting two objects which are respectively detected from two images taken at different points in time among said plurality of images and which can be considered to be a same object;

associating said plurality of objects with each other based on a result of tracking objects by said tracking step;

evaluating said each object detected with said detecting step, respectively; and

evaluating another object which has been associated by said associating step with the object to which the evaluation has been given by said evaluating objects step, based on the evaluations made by said evaluating another object step,

wherein said associating step executes association when one or more of the following conditions occur including a condition of associating objects in a case in which the objects are coupled into one object, a condition of associating of objects in a case in which one object is separated into said objects, and a condition of associating objects in a case in which distances among the objects become shorter than a predetermined threshold value.

10. The image processing method according to claim 9, further comprising a step of:

displaying said each object detected at said object detecting step in a display style representative of association of said plurality of objects.

11. The image processing method according to claim 9, wherein said predetermined judgment criterion includes whether each of said objects exists in a specific area of the image.

12. A computer program embodied on a non-transitory computer readable medium having computer readable program code means embodied in said medium for processing an image, said computer readable program code means when executed causes a computer to perform the steps of:

detecting each object of a plurality of objects contained in a plurality of images arrayed in order of points in time at which said images are taken, respectively;

tracking an object by connecting two objects which are respectively detected from two images taken at different points in time among said plurality of images and which can be considered to be a same object;

associating said plurality of objects with each other based on a result of tracking objects by said tracking step;

evaluating said each object detected with said detecting step, respectively; and

evaluating another object which has been associated by said associating step with the object to which the evaluation has been given by said evaluating objects step, based on the evaluations made by said evaluating another object step,

wherein said associating step executes association when one or more of the following conditions occur including a condition of associating objects in a case in which the plurality of objects are coupled into one object, a condition of associating objects in a case in which one object is separated into said objects, and a condition of associating objects in a case in which distances among the objects become shorter than a predetermined threshold.