Image processing apparatus having apparatus for correcting the image processing

• US 5,182,776 A
• Filed: 03/04/1991
• Issued: 01/26/1993
• Est. Priority Date: 03/02/1990
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Kasai, Syoji, Yamazaki, Masachika, Suzuki, Masato, Yoneda, Kenji, Takenaga, Hiroshi, Sakai, Yoshio, Nakamura, Kiyoshi, Inaba, Hiromi, Nakata, Naofumi, Oonuma, Naoto
• Primary Examiner(s): Couso, Jose L.
• Application Number:
  US07/664,319
• Time in Patent Office: 694 Days
• Field of Search: 382/14, 382/15, 382/34, 382/34, 382/41, 382/48, 382/51, 364/222, 358/108, 358/93, 187/132, 187/139, 340/825.28, 340/825.31
• US Class Current: 382/100
• CPC Class Codes: B66B 1/2458 For elevator systems with multiple shafts and a single car per shaft
• CPC Class Codes: B66B 1/34 Details , e.g. call counting devices, data transmission from car to control system, devices giving information to the control system
• CPC Class Codes: B66B 1/3415 Control system configuration and the data transmission or communication within the control system
• CPC Class Codes: B66B 1/3476 Load weighing or car passenger counting devices B66B5/14 takes precedence
• CPC Class Codes: B66B 2201/102 Up or down call input
• CPC Class Codes: B66B 2201/222 Taking into account the number of passengers present in the elevator car to be allocated
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes

Abstract

An image of an elevator hall or an inside of elevator car is acquired by photographing apparatus, and the number of waiting passenger is detected by comparing the above-mentioned image with a background image when no passenger is present at the elevator hall. Second image processing apparatus with high precision is prepared by the same image information as that of first image processing apparatus, teacher information derived therefrom is compared with an output from the first image processing apparatus. If the teacher information is not coincident with the output from the first image processing means, parameters required for performing the image process by the first image processing appartus, for instance, constants, threshold values and weight coefficients employed in an image processing algorithm are adjusted.

Claims

1. An image processing apparatus wherein image information is inputted to image processing means which performs an image processing on said image information based on a predetermined image processing algorithm, comprising:

means for generating teacher information related to a result of said image processing from said image processing means; and

means for correcting either said image processing algorithm, or a parameter thereof based upon said teacher information;

wherein said image processing means includes means, supplied with said image information, for performing an image processing to recognize the number of an object, and said teacher information generating means includes means for detecting a weighted value of said object, and means for computing teacher information related to the number of an object based on the detected weighted value.

2. An image processing apparatus as claimed in claim 1, wherein said teacher information generating means is supplied with image information identical to the image to be processed in said image processing means and includes means for executing an image processing with higher precision than that of said image processing means.

3. An image processing apparatus according to claim 1

wherein said teacher information generating means generates said teacher information related to an image processing result from said image processing means with a delay, as compared with said image processing performed by said image processing means.

4. An image processing apparatus according to claim 1

wherein said teacher information generating means produces teacher information related to an output of said image processing means by performing a different image processing than that in said image processing means.

6. An image processing apparatus as claimed in claim 1, wherein said correcting means includes means for selecting either a plurality of algorithms or the parameters thereof which have been previously prepared.

7. An image processing apparatus according to claim 1

wherein said correcting means includes means for determining a correcting amount with respect to an adjusting parameter of said image processing algorithm based upon the teacher information.

9. An image processing apparatus according to claim 1 further comprising:

a plurality of said image processing means for receiving image information different from each other;

wherein said teacher information generating means is commonly provided with said plurality of image processing means, for selectively producing teacher information related to image-processed results from said plurality of image processing means; and

wherein said correcting means, based on the teacher information, corrects either an image processing algorithm or a parameter thereof of each of said plurality of image processing means corresponding to the teacher information.

10. An image processing apparatus as claimed in claim 9, wherein said teacher information generating means includes second image processing means capable of selectively inputting said image information different with each other.

11. An image processing apparatus according to claim 1, further comprising:

said image processing means being a neural network having multiple layers; and

leaning means for correcting a weight coefficient of said neural network based upon said teacher information.

12. An image processing apparatus as claimed in claim 11, further comprising:

means for periodically initiating said learning means.

13. An image processing apparatus as claimed in claim 7, wherein said image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images are below than a certain threshold value with respect to a predetermined continuous comparison number.

14. An image processing apparatus as claimed in claim 1, wherein the image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images are below than a certain threshold value with respect to a predetermined continuous comparison number.

15. An image processing apparatus as claimed in claim 3, wherein said image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images are below than a certain threshold value with respect to a predetermined continuous comparison number.

17. A group-supervised elevator controlling apparatus comprising:

a plurality of elevator cars for providing service to multiple floors;

destination calling means provided within said plural cars;

elevator control means for stopping the car at the floor corresponding to at least said elevator calling;

hall calling means provided at each of floors;

group supervising control means for determining the served car in response to at least the hall calling;

a plurality of photographing means provided at arbitrary floors for outputting image information;

first image processing means, provided with the corresponding floors, for selectively inputting said image information and recognizing the number of waiting passengers at the floors by performing a first image processing on said image information based on a first image processing algorithm;

means for transferring outputs of said first image processing means to said group supervising control means;

second image processing means for selectively inputting said image information and recognizing the number of waiting passengers at the relevant floors by performing a second image processing on said image information based on a second image processing algorithm different from said first image processing algorithm; and

means for correcting either said first image processing algorithm, a parameter thereof for said first image processing means based upon an output from said second image processing means.

18. A group-supervised elevator controlling apparatus according to claim 17

wherein said second image processing includes means for detecting a weighted value of a passenger;

and

wherein said correcting means includes means for correcting either said first image processing algorithm or a parameter thereof based on the detected weighted value.

19. A group-supervised elevator controlling apparatus according to claim 17 wherein said second image processing means comprises:

means for generating teacher information related to an image-processed result from said image processing means, and

means for judging a condition related to an object to be image-processed; and

wherein said correcting means includes means for correcting either said image processing algorithm or a parameter thereof based on said teacher information and an output from said condition judging means.

20. A group-supervised elevator controlling apparatus according to claim 17 further comprising:

said second image processing means including means for generating teacher information related to said result of said image processing from said first image processing means;

means for inputting both the result of said image processing from said first image processing means and the information from said group supervising control means so as to produce an initiation instruction to said teacher information generating means; and

said correcting means including means for correcting said image processing algorithm or a parameter thereof based on the teacher information.

21. A group-supervised elevator controlling apparatus as claimed in claim 19, wherein said first image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images are below than a certain threshold value with respect to a predetermined continuous comparison number.

22. A group-supervised elevator controlling apparatus as claimed in claim 20, wherein said first image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images re below than a certain athreshold value with respect to a predetermined continuous comparison number.

23. A group-supervised elevator controlling apparatus as claimed in claim 17, wherein at least one of said first image processing means and second image processing means includes:

judging means for judging whether or not a waiting passenger is present at an elevator hall based on elevator information derived from either the group supervising control means, or the elevator control means; and

background image forming means for forming a background image in response to an output from said judging means.

24. A group-supervised elevator controlling apparatus as claimed in claim 17, wherein said background image forming means fetches a plurality of hall images in a time sequential mode, successively compares two adjoining hall images with each other, and also determines one hall image as a background image in case that variations in the hall images is below than a certain threshold value with respect to a predetermined continuous comparison number.

25. A group-supervised elevator controlling apparatus as claimed in claim 17, wherein the elevator information corresponds to at least one of a state where a door of an elevator is opened/closed at the hall entrance, hall calling information, and also car position information.

26. An image processing apparatus as claimed in claim 24, wherein the elevator information corresponds to at least one of a state where a door of an elevator is opened/closed at the hall entrance, hall calling information, and also car position information.

28. A group-supervised elevator controlling apparatus as claimed in claim 17, wherein said correcting means includes means for selecting either a plurality of algorithms or the parameters thereof which have bene previously prepared.

27. An image processing apparatus wherein image information is inputted to first image processing means which performs an image processing on said image information based on a predetermined image processing algorithm, comprising:

means for generating teacher information related to a result of said image processing from said image processing means; and

means for correcting either said image processing algorithm, or a parameter thereof based upon said teacher information;

a plurality of elevator cars for providing service to multiple floors;

destination calling means provided within said plural cars;

elevator control means for stopping the car at the floor corresponding to at least said elevator calling;

hall calling means provided at each of floors;

group supervising control means for determining the served car in response to at least the hall calling;

a plurality of photographing means provided at arbitrary floors for outputting image information;

said first image processing means being provided with the corresponding floors, for selectively inputting said image information and recognizing the number of waiting passengers on the floors by performing a first image processing on said image information based on a first image process algorithm;

means for transferring outputs of said first image processing means to said group supervising control means;

said teacher information generating means being second image processing means for selectively inputting said image information and recognizing the number of waiting passengers at the relevant floors by performing a second image processing on said image information based on a second image processing algorithm different from said first image processing algorithm; and

said means for correcting corrects either said first image processing algorithm, or a parameter thereof for said first image processing means based upon an output from said second image processing means.

5. An image processing apparatus according to claim 27

wherein said teacher information generating means includes means for judging a condition related to an image object to be processed;

and

said correcting means includes means for correcting either said image processing algorithm, or a parameter thereof in response to an output from said judging means.

8. An image processing apparatus as claimed in claim 5, wherein said image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images are below than a certain threshold value with respect to a predetermined continuous comparison number.

16. A group-supervised elevator controlling apparatus as claimed in claim 8, wherein said first image processing means includes background image forming means;

and said background image forming means successively compares two adjoining images of image information which are inputted in a time sequential mode with each other, and determines one image as a background image in case that variations in the images are below than a certain threshold value with respect to a predetermined continuous comparison number.