Method and apparatus for detecting corresponding regions between picture images

• US 4,962,540 A
• Filed: 11/27/1989
• Issued: 10/09/1990
• Est. Priority Date: 09/26/1986
• Status: Expired due to Term

• Current Assignee: Olympus Optical Corporation Limited (Olympus Corporation)
• Original Assignee: Olympus Optical Corporation Limited (Olympus Corporation)
• Inventors: Tsujiuchi, Junpei, Kikuchi, Susumu, Ohyama, Nagaaki, Honda, Toshio, Badique, Eric, Ikeda, Shigeto
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Couso, Jose L.
• Application Number:
  US07/453,382
• Time in Patent Office: 316 Days
• Field of Search: 364/526, 382/1, 382/17, 382/18, 358/27, 358/28, 358/29, 358/29 C, 358/31, 358/80
• US Class Current: 382/165
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction

Abstract

The method of detecting corresponding regions between primary color picture images comprises respective steps of converting three primary color signals of color picture images to at least two signals based on these three primary color signals, calculating the variance of the mutual concentration distribution in the operation object regions of the three primary color picture images from these converted signals and detecting the set of regions in which the calculated variance value is minimum. The apparatus for realizing the above mentioned method comprises a converting device for converting three primary color picture signals of color picture images to at least two signals based on these three primary color signals, a variance calculating device for calculating the variance of the mutual concentration distribution in the operation object regions of the three primary color picture images from the signals converted by this converting device and a corresponding region detecting device for detecting the set of regions in which the variance value calculated by this variance calculating device is minimum.

Claims

1. A method for detecting corresponding regions between primary color picture images, comprising respective steps of:

detecting mutual color concentration distribution in operation object regions;

converting three primary color signals of color picture images to at least two signals based on said three primary color signals so as to correct influence on brightness of said color picture images;

calculating the variance of the mutual color concentration distribution in the operation object regions of said three primary color picture images from at least two signals obtained by converting said three primary color signals according to said step of converting; and

thereafter

detecting a set of said operation object regions of said three primary color picture images in which the variance value which is calculated according to said step of calculating is minimum and which is the corresponding region.

3. A method according to claim 1 or 2 wherein said step of converting three primary color signals to at least two signals based on said three primary color signals includes converting the same by dividing the respective concentrations of the picture elements located in same positions of the operation object regions of the three primary color picture images respectively by the sume of the respective concentrations.

4. A method according to claim 3 wherein said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of the distance between coordinate of each mutual concentration value as converted and the coordinate of an average of the concentration values as converted.

5. A method according to claim 1 or 2 wherein said step of converting said three primary color signals to at least two signals based on said three primary color signals includes converting the same by dividing the respective concentrations of the picture elements located in same positions of the operation object regions of two picture images among the three primary color picture images by the sum of the concentrations of the picture elements located in the same positions of the operation object regions of the three primary color picture images.

6. A method according to claim 5 wherein said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of a distribution of the concentration differences of the picture elements located in the same positions of the operation object regions of said two picture images as converted.

7. A method according to claim 1 or 2 wherein said step of converting said three primary color signals to at least two signals based on said three primary color signals includes converting said three primary color signals to three signals including a signal representing the brightness and said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images.

8. A method according to claim 7 wherein said three signals including the signal representing the brightness are a brightness signal Y signal and two of a chromatic aberration signal I signal and Q signal.

9. A method according to claim 7 wherein said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of the distance between the coordinate of each of two signals corresponding to the respective picture images and the coordinate of the average of said two signals.

10. A method according to claim 1 or 2 wherein said step of converting said three primary color signals to at least two signals based on said three primary color signals includes detecting a principal component direction in which the variance of a mutual concentration distribution of said three primary color picture images is largest and two directions intersecting at right angles with it and converting said three primary color signals to be of three component values including a principal component value in which the variance is largest and said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of the mutual concentration distribution in the operation object regions of said three color picture images from two values obtained by excluding the principal component value in which the variance is largest from said three principal values.

11. A method according to claim 10 wherein said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of the distance between a coordinate of each of said two values corresponding to respective picture elements and a coordinate of the average of said two values in a plane in which said two values obtained by excluding the principal component value in which the variance is largest from said three principal component values are respectively made variable.

12. A method according to claim 1 or 2 wherein said step of converting said three primary color signals to at least two signals based on said three primary color signals includes converting the same by calculating the standard deviation of concentration values of respective picture elements in the operation object regions of said three primary color picture images and dividing the concentrations of the respective picture elements of the operation object regions of the respective picture images by said standard deviation in said picture images.

13. A method according to claim 12 wherein said step of calculating the variance of the mutual concentration deviation in the operation object regions of said three primary color picture images includes calculating the variance of the distance between a coordinate of each mutual concentration value as converted and the coordinate of the average of concentration values as converted in a space in which the concentrations of the picture elements located in same positions of the operation object regions of said three primary color picture images are respectively made variables.

14. A method according to claim 1 or 2 wherein said step of converting said three primary color signals to at least two signals based on said three primary color signals includes converting the same by calculating respectively the standard deviations of concentration values of respective picture elements in the operation object regions of said three primary color picture images and dividing respectively the concentrations of the respective picture elements of the operation object regions of respective picture images by said standard deviation in said picture image and further by the sum of concentrations divided by said standard deviation.

15. A method according to claim 14 wherein said step of calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images includes calculating the variance of the distance between a coordinate of each mutual concentration value as converted and a coordinate of the average of concentration values as converted in a space in which the concentrations of the picture elements located in same positions of the operation object regions of said three primary color picture images are respectively made variables.

2. A method for detecting corresponding regions among three primary color picture images which are correlated with a form distribution in an actual field, comprising respective steps of:

detecting mutual color concentration distribution in operation object regions;

memorizing three primary color picture images in respective picture image memorizing means;

setting operation object regions for said respective picture images which have been memorized in said picture image memorizing means;

converting three primary color signals in the operation object regions of the respective picture images to at least two signals based on said three primary color signals so as to correct influence on brightness of said color picture images;

calculating the variance of the mutual color concentration distribution in the operation object regions of said three primary color picture images from at least two signals obtained by converting said three primary color signals according to said step of converting; and

thereafter

comparing the respective variance values in a plurality of sets of operation object regions of said three primary color picture images which are obtained by changing one of said operation object regions and detecting the set of regions in which the variance value which is calculated according to said step of calculating is minimum and which is the corresponding region.

16. A method for correcting color displacements of electronic endoscope picture images by a field sequential system, comprising respective steps of:

detecting mutual color concentration distribution in operation object regions;

setting first and second operation object regions with each picture element as a center for the respective picture elements for two respective picture images of three primary color picture images;

setting a third operation object region for detecting the regions corresponding to said first and second operation object regions for another picture image among said three primary color picture images;

converting said three primary color signals in said operation object regions of the respective picture images to at least two signals based on said three primary color signals;

calculating the variance of the mutual color concentration distribution in said first, second and third operation object regions from the converted signals;

comparing the respective variance values in a plurality of sets of the operation object regions obtained by converting said third operation object region and detecting the set of regions in which the variance value is at a minimum;

detecting the displacement between the regions in which the variance value is at a minimum; and

thereafter

correcting an address of the picture element in the center of said third operation object region when the variance value is at a minimum.

17. An apparatus for detecting corresponding regions between primary color picture images, comprising:

detecting means for detecting mutual color concentration distribution in operation object regions;

a converting means for converting three primary color signals of color picture images to at least two signals based on said three primary color signals so as to correct influence on brightness of said color picture images;

a variance calculating means operably connected to said converting means for calculating the variance of the mutual color concentration distribution in the operation object regions of said three primary color picture images from at least two signals obtained by converting said three primary color signals according to said converting means; and

a corresponding region detecting means operably connected to said variance calculating means for detecting the set of operation object regions of said three primary color picture images in which the variance value calculated by said variance calculating means is minimum and which is the corresponding region.

19. An apparatus according to claim 17 or 18 wherein said converting means comprises a calculating means for converting the same by dividing the respective concentrations of the picture elements located in same positions of the operation object regions of said three primary color picture images by the sum of the respective concentrations.

20. An apparatus according to claim 19 wherein said variance calculating means comprises a calculating means for calculating the variance of the distance between a coordinate of each mutual concentration value as converted and a coordinate of the average of concentration values as converted in a space in which the concentrations of the picture elements located in the same positions of the operation object regions of said three primary color picture images are respectively made variables.

21. An apparatus according to claim 19 wherein said variance calculating means comprises:

a first calculating means for calculating the square of the deviation between a concentration as converted of each picture element and the average value of the concentrations as converted;

a second calculating means for adding the calculation results of said first calculating means of respective operation object regions; and

a third calculating means for calculating the variance of the square root of the calculation result of said second calculating means.

22. An apparatus according to claim 21 wherein said third calculating means comprises a fourth calculating means for calculating the average value of the calculation results of said second calculating means, a fifth calculating means for calculating the square of the average of the square roots of said second calculation results and a sixth calculating means for substrating said fifth calculation result from said fourth calculation result and calculating the variance of the square roots of the calculation results of said second calculating means.

23. An apparatus according to claim 17 or 18 wherein said converting means comprises a calculating means for converting respective concentrations of the picture elements located in same positions of the operation object regions of the two picture images of said three primary color picture images by dividing the same by the sum of the concentrations of the picture elements located in same positions of the operation object regions of said three primary color picture images.

24. An apparatus according to claim 23 wherein said variance calculating means comprises a calculating means for calculating the variance of a distribution of the differences of concentrations of the picture elements located in the same positions of the operation object regions of said two picture images as converted.

25. An apparatus according to claim 17 or 18 wherein said converting means comprises a means for converting three primary color signals to three signals including a signal representing the brightness and said variance calculating means comprises a calculating means for calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images from the two signals obtained by excluding the signal representing the brightness from said three signals as converted.

26. An apparatus according to claim 25 wherein said converting means is a means for converting said three primary color signals to a brightness signal Y signal and two chromatic aberration signals I signal and Q signal.

27. An apparatus according to claim 25 wherein said variance calculating means comprises a calculating means for calculating the variance of the distance between a coordinate of each of the two signals corresponding to respective picture elements and a coordinate of the average of the two signals in a plane in which the two signals obtained by excluding the signal representing the brightness from the three signals as converted as respectively made variables.

28. An apparatus according to claim 17 or 18 wherein said converting means comprises a principal component analyzing means for detecting the principal component direction in which the variance of the mutual concentration distribution of the three primary color picture images is largest and two directions intersecting at right angles with said principal component direction and a converting means for converting the three primary color signals to be of the three principal component values including a principal component value in which the variance is largest and said variance calculating means comprises a calculating means for calculating the variance of the mutual concentration distribution in the operation object regions of said three primary color picture images from the two values obtained by excluding a primary component value in which the variance is largest from said three principal component values.

29. An apparatus according to claim 28 wherein said variance calculating means comprises a calculating means for calculating the variance of the distance between a coordinate of each of the two values corresponding to the respective picture elements and a coordinate of the average of the two values.

30. An apparatus according to claim 17 or 18 wherein said converting means comprises a calculating means for calculating respectively the standard deviations of concentration values for the respective picture images in the operation object regions of said three primary color picture images and a converting means for converting the concentrations of the operation object regions of the respective picture images by dividing the same by said standard deviation in said picture images.

31. An apparatus according to claim 30 wherein said variance calculating means comprises a calculating means for calculating the variance of the distance between a coordinate of each mutual concentration value as converted and the coordinate of the average of the concentration values as converted in a space in which concentrations of the picture elements located in same positions of the operation object regions of said three primary color picture images are respectively made variables.

32. An apparatus according to claim 17 or 18 wherein said converting means comprises a calculating means for calculating respectively the standard deviations of concentration values for respective picture elements in the operation object regions of said three primary color picture images, a calculating means for dividing the concentrations of the respective picture elements of the operation object regions of the respective picture images respectively by said standard deviation in said picture images and a converting means for converting concentrations of the respective picture elements divided by said standard deviation by dividing the same by the sum of the concentrations divided by said standard deviation.

33. An apparatus according to claim 32 wherein said variance calculating means for calculating the variance of the distance between a coordinate of each mutual concentration value as converted and a coordinate of the average of concentration values as converted in a space in which concentrations of the picture elements located in same positions of the operation object regions of said three primary color picture images are respectively made variables.

18. An apparatus for detecting corresponding regions between picture images, comprising:

detecting means for detecting mutual color concentration distribution in operation object regions;

a picture image memorizing means for memorizing respective three primary color picture images;

a region setting means operably connected to said picture image memorizing means for setting respective operation object regions for the respective picture images memorized in said picture image memorizing means;

a converting means operably connected to said region setting means for converting said three primary color signals in the operation objects regions of the respective picture images to at least two signals based on said three primary color signals so as to correct influence on brightness of said color picture images;

a variance calculating means operably connected to said converting means for calculating the variance of the mutual color concentration distribution in the operation object regions of said three primary color picture images from at least two signals obtained by converting said three primary color signals according to said converting means;

a comparing means operably connected to said variance calculating means for comparing the variance value calculated by said variance calculating means with a reference value;

a memorizing means capable of memorizing said reference value and an address of the operation object region;

a memory controlling means for controlling said memorizing means to memorize said variance value as a new reference value in case the output of said comparing means is input and said variance value is smaller than said reference value and the address of the operation object region then becomes a new address; and

a controlling means for comparing said variance value with said reference value by said comparing means for obtaining a variance value which is calculated according to said variance calculating means in a different set of operation object regions of said three primary color picture images and then for controlling said region setting means to change one of said operation object regions.

34. An apparatus for correcting color displacements of electronic endoscope picture images by a field sequential system, comprising:

detecting means for detecting mutual color concentration distribution in operation object regions;

a region setting means for setting first and second operation object regions with the picture images as centers for the respective picture images for two respective picture images of the three primary color picture images, and for setting a third operation object region for detecting regions corresponding to said first and second operation object regions for another picture image of said three primary color picture images;

a converting means operably connected to said region setting means for converting the three primary color signals in said operation object regions of the respective picture images to at least two signals based on said three primary color signals;

a variance calculating means operably connected to said converting means for calculating the variance of the mutual color concentration distribution in the first, second and third operation object regions set by said region setting means from the signals converted by said converting means;

a corresponding region detecting means operably connected to a variance calculating means for comparing the respective variance values in a plurality of sets of the operation object regions obtained by changing the third operation object region by said region setting means and detecting the set of regions in which the variance value is at a minimum;

a detecting means operably connected to said corresponding region detecting means for detecting the displacement between the regions in which the variance value is at a minimum and which are detected by said corresponding region detecting means;

a memorizing means operably coupled to said detecting means for memorizing the displacement detected by said detecting means; and

an address corresting means operably connected to said memorizing means for correcting an address of the picture element in the center of the third operation object region when the variance value is at a minimum on the basis of the displacement memorized in said memorizing means.

35. An electronic endoscope apparatus of a field sequential system, comprising:

detecting means for detecting mutual color concentration distribution in operation object regions;

an endoscope having an illuminating means for sequentially radiating lights of different colors onto an object to be imaged, and an imaging means for imaging said object image corresponding to the illuminating lights of the respective colors radiated by said illuminating means;

a first memorizing means operably connected to said imaging means for memorizing three respective primary color picture images imaged by said imaging means;

a region setting means operably connected to said first memorizing means for setting respective operation object regions for said three primary color picture images memorized in said first memorizing means;

a converting means operably connected to said region setting means for converting the three primary color signals in the operation object regions of the respective picture images to at least two signals based on said three primary color signals;

a variance calculating means operably connected to said converting means for calculating the variance of the mutual color concentration distribution in the operation object regions of said three primary color picture images from the signals converted by said converting means;

a corresponding region detecting means operably connected to said variance calculating means for detecting the set of regions in which the variance value calculated by said variance calculating means is at a minimum;

a detecting means operably connected to said corresponding region detecting means for detecting the displacement between the regions in which the variance value detected by said corresponding region detecting means is at a minimum;

an address correcting means operably connected to said detecting means for correcting addresses of said three primary color picture images on the basis of the displacement detected by said detecting means;

a second memorizing means operably connected to said address correcting means for memorizing the picture images having the corrected addresses; and

a signal processing means operably connected to said first and second memorizing means for forming video signals having color displacements corrected by reading the picture image having the corrected address out of said second memorizing means, and for reading the picture image having the uncorrected address out of said first memorizing means.