Auto-focusing system and method

Patent metadata

Field	Value
Publication number	US-9426351-B2
Application number	US-201414548745-A
Country	US
Kind code	B2
Filing date	Nov 20, 2014
Priority date	Jul 10, 2014
Publication date	Aug 23, 2016
Grant date	Aug 23, 2016

Abstract

An auto-focusing method includes dividing an image into a plurality of blocks, performing discrete cosine transform (DCT) on image data of the plurality of blocks to output a plurality of DCT blocks each comprising DCT coefficients, generating a DCT mask for selecting DCT coefficients corresponding to a selected frequency range in a DCT block among the plurality of DCT blocks, and calculating a focus value of the image by applying the generated DCT mask to the DCT block.

First claim

What is claimed is:

1. An auto-focusing method comprising:
dividing an image into a plurality of blocks;
performing discrete cosine transform (DCT) on image data of the plurality of blocks to output a plurality of DCT blocks each comprising DCT coefficients;
generating a DCT mask for selecting DCT coefficients corresponding to a selected frequency range in a DCT block among the plurality of DCT blocks; and
calculating a focus value of the image by applying the DCT mask to the DCT block;
wherein the DCT block comprises a plurality of alternating current (AC) components divided into at least one low frequency component, at least one medium frequency component, and at least one high frequency component, and
wherein the generating the DCT mask comprises generating:
a first DCT mask for selecting DCT coefficients corresponding to the low frequency component and a first medium frequency component adjacent to the low frequency component; and
a second DCT mask for selecting DCT coefficients corresponding to the high frequency component and a second medium frequency component adjacent to the high frequency component.

2. The auto-focusing method of claim 1 , wherein the calculating the focus value comprises:
calculating a first focus value of the image by applying the first DCT mask to the DCT block;
calculating a second focus value of the image by applying the second DCT mask to the DCT block; and
selecting one of the first focus value and the second focus value.

3. The auto-focusing method of claim 2 , wherein the selecting comprises selecting one of the first focus value and the second focus value which has a lager change rate of a focus value.

4. The auto-focusing method of claim 2 , wherein the selecting comprises:
selecting the first focus value if the second focus value is less than a threshold value; and
selecting the second focus value if the second focus value is equal to or greater than the threshold value.

5. The auto-focusing method of claim 1 , wherein the generating the DCT mask comprises:
generating the first DCT mask if the focus value is less than a threshold value; and
generating the second DCT mask if the focus value is equal to or greater than the threshold value.

6. The auto-focusing method of claim 1 , wherein the high frequency component comprises a first high frequency component and a second high frequency component that is higher than the first high frequency component, and
wherein the second DCT mask is used for selecting DCT coefficients corresponding to the first high frequency component and the second medium frequency component adjacent to the first high frequency component.

7. The auto-focusing method of claim 1 , wherein weights are applied differently to the DCT coefficients selected by the first DCT mask or the second DCT mask.

8. The auto-focusing method of claim 7 , wherein a weight of the first DCT mask applied to the low frequency component increases in inverse proportion to a frequency value, and
wherein a weight of the second DCT mask applied to the high frequency component increases in proportion to the frequency value.

9. The auto-focusing method of claim 1 , wherein the calculating the focus value comprises:
calculating an average of absolute values of DCT coefficients of each DCT block that is masked with the DCT mask; and
calculating an average of the averages of the absolute values of the DCT coefficients calculated with respect to the plurality of DCT blocks of the image, as the focus value of the image.

10. An auto-focusing system comprising:
a divider configured to divide an image into a plurality of blocks;
a discrete cosine transform (DCT) processor configured to perform DCT on image data of the plurality of blocks to output a plurality of DCT blocks each comprising DCT coefficients;
a mask generator configured to generate a DCT mask for selecting DCT coefficients corresponding to a selected frequency range in a DCT block among the plurality of DCT blocks; and
a focus value calculator configured to calculate a focus value of the image by applying the DCT mask to the DCT block,
wherein the mask generator is configured to divide a plurality of alternating current (AC) components of the DCT block into at least one low frequency component, at least one medium frequency component, and at least one high frequency component, and
wherein the mask generator is further configured to generate:
a first DCT mask for selecting DCT coefficients corresponding to the low frequency component and a first the medium frequency component adjacent to the low frequency component; and
a second DCT mask for selecting DCT coefficients corresponding to the high frequency component and a second medium frequency component adjacent to the high frequency component.

11. The auto-focusing system of claim 10 , wherein the focus value calculator comprises:
a first calculator configured to calculate a first focus value of the image by applying the first DCT mask to the DCT block;
a second calculator configured to calculate a second focus value of the image by applying the second DCT mask to the DCT block; and
a selector configured to select one of the first focus value and the second focus value.

12. The auto-focusing system of claim 11 , wherein the selector is configured to select one of the first focus value and the second focus value which has a lager change rate of a focus value.

13. The auto-focusing system of claim 11 , wherein the selector is configured to select the first focus value if the second focus value is less than a threshold value, and select the second focus value if the second focus value is equal to or greater than the threshold value.

14. The auto-focusing system of claim 10 , wherein the mask generator is configured to generate the first DCT mask if the focus value is less than a threshold value, and generate the second DCT mask if the focus value is equal to or greater than the threshold value.

15. The auto-focusing system of claim 10 , wherein the high frequency component comprises a first high frequency component and a second high frequency component that is higher than the first high frequency component, and
wherein the second DCT mask is used for selecting DCT coefficients corresponding to the first high frequency component and the second medium frequency component adjacent to the first high frequency component.

16. The auto-focusing system of claim 10 , wherein the mask generator is further configured to apply weights differently to the DCT coefficients selected by the first DCT mask or the second DCT mask.

17. The auto-focusing system of claim 16 , wherein a weight of the first DCT mask applied to the low frequency component increases in inverse proportion to a frequency value, and
wherein a weight of the second DCT mask applied to the high frequency component increases in proportion to the frequency value.

18. The auto-focusing system of claim 10 , wherein the focus value calculator is configured to calculate an average of absolute values of DCT coefficients of each DCT block that is masked with the DCT mask, and calculate an average of the averages of the absolute values of the DCT coefficients calculated with respect to the plurality of DCT blocks of the image, as the focus value of the image.

Assignees

• Hanwha Techwin Co Ltd

Inventors

• Lee Kyungeun
• Park Youngseo
• Park Dongseuk
• Lee Chunwon
• Kwon Youngsang
• Kim Daebong

Classifications

• H04N23/673 Primary

  based on contrast or high frequency components of image signals, e.g. hill climbing method · CPC title

• G06V10/431

  Frequency domain transformation; Autocorrelation · CPC title

• G06K9/52

  Physics · mapped topic

• G06K9/46

  Physics · mapped topic

• G06K2009/4666

  Physics · mapped topic