IMAGE CONVERSION APPARATUS, METHOD, AND STORAGE MEDIUM

• US 20130223762A1
• Filed: 02/22/2013
• Published: 08/29/2013
• Est. Priority Date: 02/28/2012
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Nagamasa, Yoshinobu
• Granted Patent: US 9,020,296 B2
• Time in Patent Office: Days
• US Class Current: 382/293
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/30176 Document
• CPC Class Codes: G06T 5/00 Image enhancement or restoration
• CPC Class Codes: G06T 5/80 Geometric correction

Abstract

An image conversion apparatus calculates, based on a first value for obtaining first coordinate values in a second image before first conversion, which correspond to coordinate values of one pixel in a first image after first conversion, a second value for obtaining second coordinate values in the second image, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image. The apparatus converts the second coordinate values into third coordinate values for second conversion of converting a third image into the second image and converts the third image into the first image. In the calculation of the second value, addition or subtraction using a constant and a result of the calculation is iteratively executed for sequentially outputting values corresponding to results of multiplication of coordinate values of each pixel in the first image and the constant.

Claims

1. An image conversion apparatus comprising:

a first calculation unit for executing, based on a first value used to obtain first coordinate values in a second image before first image conversion, which correspond to coordinate values of one pixel in a first image after the first image conversion, calculation of obtaining a second value used to obtain second coordinate values in the second image before the first image conversion, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image after the first image conversion;

a second calculation unit for executing calculation of obtaining the second coordinate values using the second value obtained by the calculation of the first calculation unit;

a coordinate conversion unit for executing coordinate conversion of converting the second coordinate values obtained by the second calculation unit into third coordinate values for second image conversion of converting a third image into the second image; and

an image conversion unit for executing the first and second image conversion of converting the third image into the first image, based on the third coordinate values converted by the coordinate conversion unit,

wherein the first calculation unit iteratively executes one of addition and subtraction using a predetermined constant and an output of the first calculation unit as inputs so as to sequentially output a value corresponding to a multiplication result obtained by multiplying coordinate values of each of a plurality of adjacent pixels in the first image after the first image conversion by the predetermined constant.

2. The apparatus according to claim 1, wherein the coordinate conversion unit nonlinearly transforms the second coordinate values into the third coordinate values.

3. The apparatus according to claim 1, wherein the coordinate conversion unit converts the second coordinate values into the third coordinate values in accordance with a distance between a predetermined position and a position indicated by the second coordinate values.

4. The apparatus according to claim 1, wherein the coordinate conversion unit converts the second coordinate values into the third coordinate value to correct a lens distortion.

5. The apparatus according to claim 1, wherein the second calculation unit obtains the second coordinate values in the second image before the first image conversion, which correspond to the coordinate values of the adjacent pixel in the first image after the first image conversion, to convert four vertices before the first image conversion into four vertices after the first image conversion.

6. The apparatus according to claim 1, wherein the first calculation unit iteratively executes, for each line, calculation of sequentially obtaining a value used to obtain coordinate values in the second image before the first image conversion, which correspond to the coordinate values of each of the plurality of adjacent pixels on one line in the first image after the first image conversion.

7. The apparatus according to claim 1, wherein the first calculation unit sequentially obtains, for each of a plurality of blocks obtained by dividing the first image after the first image conversion, a value used to obtain coordinate values in the second image before the first image conversion, which correspond to the coordinate values of each of the plurality of adjacent pixels in the first image after the first image conversion.

8. An image conversion apparatus comprising:

a calculation unit for executing, based on a first value used to obtain coordinate values in a second image before image conversion, which correspond to coordinate values of one pixel in a first image after the image conversion, calculation of obtaining a second value used to obtain coordinate values in the second image before the image conversion, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image after the image conversion; and

a conversion unit for converting an image to be converted, based on the second value obtained by the calculation unit,

wherein the calculation unit iteratively executes, for each of a plurality of blocks obtained by dividing the first image after the image conversion, one of addition and subtraction using a predetermined constant and an output of the calculation unit as inputs so as to sequentially output a value corresponding to a multiplication result obtained by multiplying coordinate values of each of a plurality of adjacent pixels in the first image after the image conversion by the predetermined constant.

9. The apparatus according to claim 8, wherein the calculation unit executes the calculation of obtaining the value used to obtain the coordinate values in the second image before the image conversion, which correspond to the coordinate values of the adjacent pixel in the first image after the image conversion, to convert four vertices before the image conversion into four vertices after the image conversion.

18. The apparatus according to claim 8, wherein the calculation unit holds a product of a predetermined coordinate element and a matrix element of an inverse matrix of a conversion matrix as an initial value, wherein the conversion matrix correlates coordinate values of one pixel in the second image before the image conversion with coordinate values in the first image after the image conversion, and

wherein the calculation unit starts calculation of the multiplication result from coordinate values including the predetermined coordinate element in the first image after the image conversion and sequentially calculates, as the multiplication result, a product of a coordinate element of one pixel in the first image after the image conversion and a matrix element of the inverse matrix.

19. The apparatus according to claim 18, further comprising a shifter for bit shifting an input value to the left by a predetermined natural number of bits,

wherein the calculation unit sets the predetermined coordinate element and the initial value for each of the plurality of blocks that has a width or a height, of which the number of pixels is a power of a number having 2 as a base and a natural number as an exponent,

wherein the predetermined coordinate element set for each of the plurality of blocks has a value obtained by multiplying a power of a number having 2 as a base and a natural number as an exponent by an integer value of 0 or more which is designated by a position of the block, and

wherein the initial value set for each of the plurality of blocks is a value obtained by the shifter bit shifting a value obtained by multiplying the matrix element by the integer value.

20. The apparatus according to claim 19, further comprising a second calculation unit for adding the matrix element to or subtracting the matrix element from a value obtained for the block by multiplying the matrix element by the integer value,

wherein the shifter calculates the initial value for another block adjacent to the block by bit shifting a calculation result of the second calculation unit to the left by the predetermined natural number of bits.

10. An image conversion method of an image conversion apparatus, comprising:

executing, based on a first value used to obtain first coordinate values in a second image before first image conversion, which correspond to coordinate values of one pixel in a first image after the first image conversion, first calculation of obtaining a second value used to obtain second coordinate values in the second image before the first image conversion, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image after the first image conversion;

executing second calculation of obtaining the second coordinate values using the second value obtained by the first calculation;

executing coordinate conversion of converting the second coordinate values obtained by the second calculation into third coordinate values for second image conversion of converting a third image into the second image; and

executing the first and second image conversion of converting the third image into the first image, based on the third coordinate values converted by the coordinate conversion,

wherein in the first calculation, one of addition and subtraction using a predetermined constant and a result of the first calculation as inputs is iteratively executed so as to sequentially output a value corresponding to a multiplication result obtained by multiplying coordinate values of each of a plurality of adjacent pixels in the first image after the first image conversion by the predetermined constant.

11. The method according to claim 10, wherein in the coordinate conversion, the second coordinate values are nonlinearly transformed into the third coordinate values.

12. The method according to claim 10, wherein in the coordinate conversion, the second coordinate values are converted into the third coordinate values in accordance with a distance between a predetermined position and a position indicated by the second coordinate values.

13. The method according to claim 10, wherein in the coordinate conversion, the second coordinate values are converted into the third coordinate values to correct a lens distortion.

14. The method according to claim 10, wherein in the second calculation, the second coordinate values in the second image before the first image conversion, which correspond to the coordinate values of the adjacent pixel in the first image after the first image conversion, are obtained to convert four vertices before the first image conversion into four vertices after the first image conversion.

15. An image conversion method of an image conversion apparatus, comprising:

executing, based on a first value used to obtain coordinate values in a second image before image conversion, which correspond to coordinate values of one pixel in a first image after the image conversion, calculation of obtaining a second value used to obtain coordinate values in the second image before the image conversion, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image after the image conversion; and

converting an image to be converted, based on the second value obtained by the calculation,

wherein in the calculation, one of addition and subtraction using a predetermined constant and an output of the calculation as inputs is iteratively executed for each of a plurality of blocks obtained by dividing the first image after the image conversion so as to sequentially output a value corresponding to a multiplication result obtained by multiplying coordinate values of each of a plurality of adjacent pixels in the first image after the image conversion by the predetermined constant.

16. The method according to claim 15, wherein in the calculation, the calculation of obtaining the value used to obtain the coordinate values in the second image before the image conversion, which correspond to the coordinate values of the adjacent pixel in the first image after the image conversion, is executed to convert four vertices before the image conversion into four vertices after the image conversion.

17. A storage medium storing a computer program, the computer program comprising:

executing, based on a first value used to obtain first coordinate values in a second image before first image conversion, which correspond to coordinate values of one pixel in a first image after the first image conversion, first calculation of obtaining a second value used to obtain second coordinate values in the second image before the first image conversion, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image after the first image conversion;

executing second calculation of obtaining the second coordinate values using the second value obtained by the first calculation;

executing coordinate conversion of converting the second coordinate values obtained by the second calculation into third coordinate values for second image conversion of converting a third image into the second image; and

executing first and second conversion of converting the image into the first image, based on the third coordinate values converted by the coordinate conversion,

wherein in the first calculation, one of addition and subtraction using a predetermined constant and a result of the first calculation as inputs is iteratively executed so as to sequentially output a value corresponding to a multiplication result obtained by multiplying coordinate values of each of a plurality of adjacent pixels in the first image after the first image conversion by the predetermined constant.