Monitoring camera and image quality improving method

Patent metadata

Field	Value
Publication number	US-12229923-B2
Application number	US-202217580089-A
Country	US
Kind code	B2
Filing date	Jan 20, 2022
Priority date	Jan 22, 2021
Publication date	Feb 18, 2025
Grant date	Feb 18, 2025

Abstract

A monitoring camera includes an imaging unit configured to capture an image of an imaging area, and a processor configured to determine a first intensity and a second intensity to be different from each other. The first intensity indicates an intensity of a noise reduction processing executed on an attention portion in the captured image of the imaging area, and the second intensity indicates an intensity of a noise reduction processing executed on a non-attention portion in the captured image. The first intensity is lower than the second intensity. The processor is configured to execute the noise reduction processing on the attention portion based on the determined first intensity, to execute the noise reduction processing on the non-attention portion based on the determined second intensity, and to output an image after the noise reduction processing.

First claim

What is claimed is:

1. A monitoring camera, comprising:
a camera configured to capture a first image of an imaging area; and
a processor configured to detect a moving body portion and a still portion of the first image, the moving body portion including a moving body and being an attention portion in the first image of the imaging area, the still portion being a non-attention portion in the first image of the imaging area;
the processor further configured to determine a first intensity and a second intensity, the first intensity being different from the second intensity, the first intensity being of a first noise reduction processing executed on the attention portion in the first image of the imaging area, the second intensity being of a second noise reduction processing executed on the non-attention portion in the first image,
wherein the first intensity is lower than the second intensity, and
wherein the processor is configured
to execute the first noise reduction processing on the attention portion based on the first intensity,
to execute the second noise reduction processing on the non-attention portion based on the second intensity, and
to output a second image after the first noise reduction processing and the second noise reduction processing are executed, whereby the second image includes plural areas with different intensities of noise reduction processing,

wherein the processor is configured to determine at least one of the first intensity and the second intensity based on an attention level for each type of a subject indicated by the attention portion.

2. The monitoring camera according to claim 1 , further comprising:
an AI calculation processor configured to detect the attention portion on an image of the imaging area using an artificial intelligence (AI) learning model.

3. The monitoring camera according to claim 2 ,
wherein the processor is configured to determine at least one of the first intensity and the second intensity based on a presence or an absence of the attention portion, an attention coefficient for each type of a subject indicated by the attention portion, a weight coefficient based on a size of the attention portion, and a certainty of the subject detected by the AI learning model.

4. The monitoring camera according to claim 1 ,
wherein the processor is configured to determine the second intensity to be a maximum value of noise reduction processing.

5. The monitoring camera according to claim 3 ,
wherein the processor is configured
to derive a first attention level of a current frame based on the presence or the absence of the attention portion, the attention coefficient for each type of the subject indicated by the attention portion, the weight coefficient based on the size of the attention portion, and the certainty of the subject detected by the AI learning model, and
to determine the first intensity by executing a spatial filter processing and a time filter processing using the first attention level of the current frame and a second attention level of a previous frame immediately before the current frame.

6. The monitoring camera according to claim 1 ,
wherein the processor is configured to determine the first intensity based on an attention level determined for each coordinate in the first image of the imaging area.

7. The monitoring camera according to claim 3 ,
wherein the processor is configured to further determine the first intensity based on a motion vector of the subject.

8. The monitoring camera according to claim 1 ,
wherein the moving body includes a person or a vehicle moving in the imaging area.

9. The monitoring camera according to claim 1 , wherein
when the processor detects the moving body portion and the still portion of the first image, a first value is assigned to each pixel of the first image in which an object is detected as a detection area value, and a second value is assigned to each pixel of the first image in which no object is detected as the detection area value, and
the first value is different than the second value.

10. The monitoring camera according to claim 9 , wherein
when the processor detects the moving body portion and the still portion of the first image, one of plural values is further assigned to each pixel of the first image for which the first value is assigned, based on a type of the object detected in the pixel, as an attention level value.

11. The monitoring camera according to claim 10 , wherein
when the processor detects the moving body portion and the still portion of the first image, a third value is further assigned to each pixel of the first image for which the first value is assigned, as a weight coefficient value, in response to a ratio of a first size of the object to a second size of the first image being less than a lower limit value, and a fourth value is further assigned to each pixel of the first image for which the first value is assigned, as the weight coefficient value, in response to the ratio of the first size of the object to the second size of the first image being larger than an upper limit value, and
the third value is different than the fourth value.

12. The monitoring camera according to claim 11 , wherein
the first value is same as the fourth value.

13. The monitoring camera according to claim 11 , wherein
the processor determines at least one of the first intensity or the second intensity at based on at least the detection area value, the attention level value, and the weight coefficient value of each pixel.

14. An image quality improving method to be executed by a monitoring camera, the image quality improving method comprising:
capturing a first image of an imaging area;
detecting, by a processor, a moving body portion and a still portion of the first image, the moving body portion including a moving body and being an attention portion in the first image of the imaging area, the still portion being a non-attention portion in the first image of the imaging area;
determining, by the processor, a first intensity and a second intensity, the first intensity being different from the second intensity, the first intensity being of a first noise reduction processing executed on the attention portion in the first image of the imaging area, the second intensity being of a second noise reduction processing executed on the non-attention portion in the first image;
executing the first noise reduction processing on the attention portion based on the first intensity, and executing the second noise reduction processing on the non-attention portion based on the second intensity; and
outputting a second image after the first noise reduction processing and the second noise reduction processing are executed, whereby the second image includes plural areas with different intensities of noise reduction processing,
wherein the first intensity is lower than the second intensity, and
wherein the processor is configured to determine at least one of the first intensity and the second intensity based on an attention level for each type of a subject indicated by the attention portion.

15. A non-transitory computer readable storage medium including an image quality improvement program, the image quality improvement program, when executed by a computer, causing the computer to perform an image quality improving method, the image quality improving method comprising:
capturing a first image of an imaging area;
detecting a moving body portion and a still portion of the first image, the moving body portion including a moving body and being an attention portion in the first image of the imaging area

Assignees

• Panasonic I Pro Sensing Solutions Co Ltd
• I Pro Co Ltd

Inventors

• Nozaki Hirotaka
• Ikeda Jun
• Aoki Yohei

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• H04N23/81

  for suppressing or minimising disturbance in the image signal generation · CPC title

• G06T7/215

  Motion-based segmentation · CPC title

• G06N20/00

  Machine learning · CPC title