System for automated transformation of gray level of image

• US 5,150,421 A
• Filed: 01/09/1991
• Issued: 09/22/1992
• Est. Priority Date: 01/09/1990
• Status: Expired due to Term

• Current Assignee: Hitachi Medical Corporation (Hitachi, Ltd.)
• Original Assignee: Hitachi, Ltd., Hitachi Medical Corporation (Hitachi, Ltd.)
• Inventors: Morishita, Koichi, Sato, Kazuhiro, Yokoyama, Tetsuo
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US07/639,221
• Time in Patent Office: 622 Days
• Field of Search: 382/51, 382/6, 382/18
• US Class Current: 382/169
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties

Abstract

Disclosed is an automated image gray level transformation system applied to a medical image processing work station for the purpose of dispensing with the prior art manual complex processing thereby improving the accuracy of diagnosis. The system carries out, in addition to its fundamental step of equalizing a histogram of image data of an input image, the steps of suppressing an adverse effect of a background region of the image so as to obtaining an intensity transformation function, constraining the intensity transformation function and modifying a bias value used for the function constraining.

Claims

1. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting result of processing by said image processing means, said system comprising, as part of said image processing means, means for continuously constraining an intensity transformation function, which is calculated from a cumulative intensity histogram for each input image, on the basis of the distance between said intensity tranformation function and a pre-selected reference intensity transformation function, further comprising means for suppressing an adverse effect of a background region other than an objective region of an input image during computation of said cumulative intensity histogram.

2. An automated gray level transformation system according to claim 1, wherein said adverse effect of said background region is suppressed by employing, during computation of said cumulative intensity histogram, a weight montonously outwardly decreasing from the central part of said input image.

3. An automated gray level transformation system according to claim 1, wherein said adverse effect of said background region is suppressed by employing, during computation of said cumulative intensity histogram, a weight corresponding to the intensity of said input image.

4. An automated gray level transformation system according to claim 1, wherein said adverse effect of said background region is suppressed by extracting a contour showing an abrupt change in intensity of said input image and removing data in the region outside of said contour.

5. An automated gray level transformation system according to claim 1, further comprising means for modifying, after minimization of said adverse effect of said background region and constraining of said intensity transformation function, a bias value which is selected to be large in a low intensity portion and small in a high intensity portion of said input image according to a characteristic curve of said constrained intensity transformation function.

6. An automated gray level transformation system according to claim 5, wherein said bias value can be modified in an interactive mode by a man-machine interface.

7. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting a result of processing by said image processing means, said system comprising means for equalizing a histogram of image data applied as an input from said image input means, means for constraining an intensity transformation function obtained as a result of the equalization of said histogram of the image data, and means for suppressing an adverse effect of a background region of an inputted image before histogram equalization by said histogram equalizing means, wherein said means for suppressing said adverse effect of said background region provides a weight corresponding to the distance from the central part of said inputted image.

8. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting a result of processing by said image processing means, said system comprising means for equalizing a histogram of image data applied as an input from said image input means, means for constraining an intensity transformation function obtained as a result of the equalization of said histogram of the image data, and means for suppressing an adverse effect of a background region of an inputted image before histogram equalization by said histogram equalizing means, wherein said means for suppressing said adverse effect of said background region provides a weight corresponding to the intensity of said inputted image.

9. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting a result of processing by said image processing means, said system comprising means for equalizing a histogram of image data applied as an input from said image input means, means for constraining an intensity transformation function obtained as a result of the equalization of said histogram of the image data, and means for suppressing an adverse effect of a background region of an inputted image before histogram equalization by said histogram equalizing means, wherein said means for suppressing said adverse effect of said background region provides a weight corresponding to the contour of an objective region in said inputted image.

10. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting a result of processing by said image processing means, said system comprising means for equalizing a histogram of image data applied as an input from said image input means, means for constraining an intensity transformation function obtained as a result of the equalization of said histogram of the image data, and means for suppressing an adverse effect of a background region of an inputted image before histogram equalization by said histogram equalizing means, wherein said means for suppressing said adverse effect of said background region includes means for providing a weight corresponding to the distance from the central part of said inputted image and means for providing a weight corresponding to the intensity of said inputted image.

11. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting a result of processing by said image processing means, said system comprising means for equalizing a histogram of image data applied as an input from said image input means, means for constraining an intensity transformation function obtained as a result of the equalization of said histogram of the image data, and means for suppressing an adverse effect of a background region of an inputted image before histogram equalization by said histogram equalizing means, wherein said means for suppressing said adverse effect of said background region includes means for providing a weight corresponding to the contour of distance from the central part of said inputted image and means for providing a weight corresponding to the contour of an objective region in said inputted image.

12. An automated gray level transformation system for an image processing system including image input means for inputting images, image processing means for processing inputted images and image output means for outputting a result of processing by said image processing means, said system comprising means for equalizing a histogram of image data applied as an input from said image input means, means for constraining an intensity transformation function obtained as a result of the equalization of said histogram of the image data, and means for suppressing an adverse effect of a background region of an inputted image before histogram equalization by said histogram equalizing means, wherein said means for suppressing said adverse effect of said background region includes means for providing a weight corresponding to the intensity of said inputted image and means for providing a weight corresponding to the contour of an objective region in said inputted image.