Method and apparatus for detecting an approaching object within a monitoring zone

• US 5,798,787 A
• Filed: 08/08/1996
• Issued: 08/25/1998
• Est. Priority Date: 08/11/1995
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Kasano, Akira, Yamaguchi, Hideaki
• Primary Examiner(s): Chin, Tommy P.
• Assistant Examiner(s): DIN, LUANNE PAN
• Application Number:
  US08/689,348
• Time in Patent Office: 747 Days
• Field of Search: 348/152, 348/153, 348/154, 348/118, 348/170, 348/148, 348/149, 348/119, 348/169-172, 348/161, 348/143, 348/116
• US Class Current: 348/152
• CPC Class Codes: H04N 7/188 Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position signal generation from motion picture films H04N5/253

Abstract

A method and system for detecting the approaching of a target object into a monitoring zone wherein the target object is scheduled to enter the zone is provided. The method includes providing motion image data continuously taken as an image from the monitoring zone, initially setting a plurality of monitoring areas in an image plane of the monitoring zone on a track along which the target object moves, extracting the image data of these monitoring areas from the motion image data, detecting an image variation of the respective monitoring areas through the processing of the image data of the monitoring areas for each monitoring area and deciding, based on the detection sequence of the monitoring areas corresponding to a detected image variation, whether of not the target object approaches into the monitoring zone.

Claims

1. A method for detecting approaching of a target object into a monitoring zone in which the target object is scheduled to enter, comprising the steps of:

continuously taking a picture of the monitoring zone to provide continuous moving picture data of the monitoring zone;

initially setting a plurality of monitoring areas along a track of the target object in an image of the monitoring zone and extracting image data from the monitoring areas from the moving picture data;

processing the image data of the monitoring areas for each monitoring area and detecting an image variation of each monitoring area; and

deciding that of the target object is approaching into the monitoring zone in accordance with a detection sequence of monitoring areas where an image variation is detected.

2. The method according to claim 1, wherein the deciding step comprises initially setting a scheduled variation sequence determining a sequence of the monitoring areas corresponding to the image variation detected when the target object approaches into the monitoring zone, comparing the detection sequence of the monitoring areas corresponding to the detected image variation with the scheduled variation sequence and deciding whether or not the target object approaches into the monitoring zone.

3. The method according to claim 2, wherein the deciding step includes:

deciding whether or not there occurs any backward deviated detection sequence relative to the scheduled variation sequence in the detection sequence of a plurality of monitoring areas corresponding to the detected image variation;

deciding approaching of whether or not there is at least one of the monitoring areas corresponding to the detected image variation in each of groups into which the plurality of monitoring areas are divided in accordance with the scheduled variation sequence; and

deciding approaching of the target object into the monitoring zone when the detection sequence is not backward deviated and decision is made to the effect that, in the respective groups, there is at least one of the monitoring areas corresponding to the detected image variation.

4. The method according to claim 2, wherein the deciding step includes:

finding a detection time interval between the monitoring areas of a forward/backward deviated adjacent detection sequence in a plurality of monitoring areas corresponding to the detected image variation; and

comparing, with an initially set threshold value, the detection time interval relative to all the monitoring area pairs of the forward/backward deviated detection sequence and, when all the detection time intervals are in the threshold value range, deciding the target object as approaching into the monitoring zone.

5. The method according to claim 2, wherein the deciding step includes:

finding a ratio (n/N) between a total number of initially set monitoring areas, N, and the number of those monitoring areas, n, corresponding to detected image variation;

deciding whether or not there occurs any backward deviated sequence relative to the scheduled variation sequence in the detection sequence of a plurality of monitoring areas corresponding to the detected image variation; and

deciding the target object as approaching into the monitoring zone both when decision is made to the effect that the detection sequence is not backward deviated and when the ratio (n/N) exceeds on initially determined threshold value.

6. The method according to claim 2, wherein the deciding step includes:

initially setting weight factors to the respective monitoring areas in accordance with the degrees of importance; and

deciding the target object as approaching into the monitoring zone when a cumulative value of the weight factors exceeds the predetermined threshold value.

7. The method according to claim 1, wherein the image variation detecting step includes;

preparing a luminance histogram from the image data of the monitoring areas; and

detecting the image variation of the monitoring areas with the use of the histogram of the monitoring areas.

8. The method according to claim 7, wherein the image variation detecting step includes:

finding a moment value of the monitoring area from the histogram of the monitoring area in accordance with the following equation of

space="preserve" listing-type="equation">the moment value=Σ

(hist(ii)×

ii)

where ii represents a luminance value and hist(ii) the frequence of the luminance value; and

deciding the image of the monitoring area as varying when the variation of the moment value is detected relative to the monitoring area.

9. An apparatus for detecting approaching of a target object into a monitoring zone in which the target object is scheduled to enter, comprising:

image inputting means for taking a picture of the monitoring zone to provide moving picture data continuously;

monitoring area controlling means for managing, based on coordinate information of a monitoring display for the monitoring zone, a plurality of monitoring areas in the monitoring zone on a track along which the target object moves;

means for extracting, from the moving picture data, image data of the plurality of monitoring areas managed by the monitoring area controlling means;

image variation detecting means for processing the image data of the monitoring areas for each monitoring area and for detecting an image variation of each of the monitoring areas; and

deciding means for deciding whether or not the target object approaches into the monitoring zone on a basis of a sequence of monitoring areas where an image variation is detected.

10. The apparatus according to claim 9, wherein the deciding means includes:

means for managing, with respect to the respective monitoring areas where an image variation is detected by the image variation detecting means, a real variation sequence showing an image variation detection sequence of the monitoring areas involved;

means for managing a scheduled variation sequence showing an image variation sequence of the monitoring areas involved upon approaching of the target object into the monitoring zone; and

means for comparing the real variation sequence with the scheduled variation sequence so as to detect that the target object approaches into the monitoring zone.

11. The apparatus according to claim 10, wherein the deciding means includes:

means for deciding whether or not there occurs any backward deviated sequence relative to the scheduled variation sequence in the real variation sequence of the monitoring areas corresponding to the detected image variation;

means for deciding whether or not there is at least one of monitoring areas corresponding to the detected image variation in each of group information items into which the plurality of monitoring areas are divided in accordance with the scheduled variation sequence; and

means for deciding the target object as approaching into the monitoring zone when the detection sequence is not backward deviated and decision is made to the effect that, in the respective group, there is at least one of the monitoring areas corresponding to the detected image variation.

12. The apparatus according to claim 10, where the deciding means includes:

means for finding a detection time interval between the monitoring areas of a backward/forward deviated adjacent detection sequence relative to a plurality of monitoring areas corresponding to the detected image variation; and

means for, comparing, with an initially determined threshold value, the detection time interval relative to all the monitoring area pairs of the forward/backward deviated detection sequence and, when the detection time interval is in the threshold value, deciding the target object as approaching into the monitoring zone.

13. The apparatus according to claim 10, wherein the deciding means includes:

means for finding a ratio (n/N) between a total number of monitoring areas, N, managed by the monitoring area controlling means and the number of the monitoring areas, n, corresponding to the image variation detected by the image variation detecting means;

means for deciding whether or not there occurs any backward deviated sequence relative to the scheduled variation sequence in the real variation sequence of a plurality of monitoring areas corresponding to the detected image variation; and

means for deciding the target object as approaching into the monitoring zone when decision is made to the effect that the detection sequence is backward deviated and the ratio (n/N) exceeds a predetermined threshold value.

14. The apparatus according to claim 10, wherein the deciding means includes:

means for reserving weight factors of the respective monitoring areas;

means for reading, from the reserving means, the weight factors of the monitoring areas corresponding to the detected image variation;

means for cumulatively adding the weight factors of the monitoring areas corresponding to detected image variation; and

means for deciding the target object as approaching into the monitoring area when the cumulatively added value of the weight factors exceeds the threshold value.

15. The apparatus according to claim 9, wherein the image variation detecting means includes:

means for preparing a luminance histogram from the image data of the monitoring areas; and

means for detecting the image variation of the monitoring areas with the use of the histogram of the monitoring areas.

16. The apparatus according to claim 9, wherein the image variation detecting means includes:

means for preparing a luminance histogram from the image data of the monitoring areas;

means for detecting the image variation of the monitoring areas with the use of the histogram of the monitoring areas;

means for finding a moment value of the respective monitoring area from the histogram of the monitoring areas in accordance with the following equation of

space="preserve" listing-type="equation">the moment value=Σ

(hist(ii)×

ii)

where ii represents a luminance value and hist(ii) the frequency of the luminance value ii; and

means for deciding the image of the monitoring area as varying when the variation of the moment value is detected with respect to the monitoring area.