Detection of image sharpness in frequency domain

Patent metadata

Field	Value
Publication number	US-2023401813-A1
Application number	US-202118035665-A
Country	US
Kind code	A1
Filing date	Nov 8, 2021
Priority date	Nov 9, 2020
Publication date	Dec 14, 2023
Grant date	—

Abstract

An image processing apparatus and method are provided which obtains an image captured by an image capture device and stored in a memory, extracts one or more regions of interest in the obtained image, normalizes the extracted one or more regions of interest to be at a same scale or some predefined scales, extract the frequency information of regions of interest, determines a sharpness of the obtained image by aggregating the frequency information in each of the one or more extracted regions of interest, and labels the obtained image with the determined sharpness score.

First claim

We claim:

1 . An image processing method comprising:
obtaining an image captured by an image capture device and stored in a memory; extracting one or more regions of interest in the obtained image; normalizing the extracted one or more regions of interest to be at a same scale or some predefined scales; extracting the frequency information of regions of interest; determining a sharpness of the obtained image by aggregating the frequency information in each of the one or more extracted regions of interest; and labeling the obtained image with the determined sharpness score.

2 . The image processing method of claim 1 , further comprising
generating a patch region having a predetermined size; determining a stride size based on a size of the patch region and scale achieved during normalization; moving the generated patch region according to the stride size to determine the intensity values and the frequency information in each of the one or more extracted regions.

3 . The image processing method of claim 2 , further comprising
selecting, for the generated patch, a ratio of frequencies in an X direction and a Y direction to identify a first region of the patch having lowest frequencies in both the X and Y direction to identify a second region of the patch having highest frequencies in just the X direction, just the Y direction and both the X and Y direction.

4 . The image processing method according to claim 3 ,
calculating one or more region scores using the frequency information in the first region, and one or more region scores using the frequency information in second region.

5 . The image processing method according to claim 4 ,
wherein the frequency information includes one or more of a mean value of frequency information and one or more of a maximum value of frequency information for each of the first region and the second region.

6 . The image processing method according to claim 2 , wherein generating the patch region further comprises
using a plurality scales to obtain information from neighboring pixels for the generated patch such that each scale evaluates an increased number of neighboring pixels in the generated patch.

7 . The image processing method according to claim 6 , further comprising
selecting a subset of scales from the plurality of scales; and obtaining frequency information about the image data from the pixels included in the selected subset of scales.

8 . The image processing method according to claim 6 , further comprising
using the frequency information from each of the plurality of scales to predict sharpness of the obtained image.

9 . An image processing apparatus comprising:
one or more memories storing instructions; and one or more processors that, in response to executing the stored instructions are configured to:
obtain an image captured by an image capture device and stored in a memory;
extract one or more regions of interest in the obtained image;
normalize the extracted one or more regions of interest to be at a same scale or some predefined scales;
extract the frequency information of regions of interest;
determine a sharpness of the obtained image by aggregating the frequency information in each of the one or more extracted regions of interest; and
label the obtained image with the determined sharpness score.

10 . The image processing apparatus of claim 9 , wherein execution of the instructions further configures the one or more processors to
generate a patch region having a predetermined size; determine a stride size based on a size of the patch region and scale achieved during normalization; move the generated patch region according to the stride size to determine the intensity values and the frequency information in each of the one or more extracted regions.

11 . The image processing apparatus of claim 10 , wherein execution of the instructions further configures the one or more processors to
select, for the generated patch, a ratio of frequencies in an X direction and a Y direction to identify a first region of the patch having lowest frequencies in both the X and Y direction to identify a second region of the patch having highest frequencies in just the X direction, just the Y direction and both the X and Y direction.

12 . The image processing apparatus according to claim 11 , wherein execution of the instructions further configures the one or more processors to
calculate one or more region scores using the frequency information in the first region, and one or more region scores using the frequency information in second region.

13 . The image processing apparatus according to claim 12 ,
wherein the frequency information includes one or more of a mean value of frequency information and one or more of a maximum value of frequency information for each of the first region and the second region.

14 . The image processing apparatus according to claim 10 , wherein execution of the instructions further configures the one or more processors to generate the patch region using a plurality scales to obtain information from neighboring pixels for the generated patch such that each scale evaluates an increased number of neighboring pixels in the generated patch.

15 . The image processing apparatus according to claim 14 , wherein execution of the instructions further configures the one or more processors to
select a subset of scales from the plurality of scales; and obtain frequency information about the image data from the pixels included in the selected subset of scales.

16 . The image processing apparatus according to claim 14 , wherein execution of the instructions further configures the one or more processors to
use the frequency information from each of the plurality of scales to predict sharpness of
the obtained image.

Assignees

• Canon Usa Inc

Inventors

• Cao Xiwu

Classifications

• G06V10/431 Primary

  Frequency domain transformation; Autocorrelation · CPC title

• G06V10/25

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06V10/32

  Normalisation of the pattern dimensions · CPC title

• G06V10/993

  Evaluation of the quality of the acquired pattern · CPC title

• G06V20/70

  Labelling scene content, e.g. deriving syntactic or semantic representations · CPC title