Method and device for image processing

1. An image processing device comprising:

a processor configured to process sensing data received from an image sensor; and

a controller configured to control the processor,

wherein the processor comprises:

an artificial intelligence (AI) image processing model trained in image processing through learning;

an arithmetic logic unit (ALU) configured to perform a computation involved in image processing using the AI image processing model; and

an image signal processor (ISP) configured to process an image by using an image processing filter,

wherein the controller is configured to analyze the sensing data in order to select the at least one device among the ALU and the ISP based on a quality of the sensing data and control the processor so that at least one device selected among the ALU and the ISP processes the sensing data,

wherein the controller comprise a data analyzer configured to analyze at least one type of data among the sensing data and pixel data processed by the processor, and an image quality assessor configured to assess a Quality of data by using a result of analyzing the at least one type of data, and

wherein the controller assesses noise among factors related to the quality of the data, and controls a device that is selected according to a range of the noise.

2. The image processing device according to claim 1 , further comprising an image sensor configured to generate the sensing data by converting light into an electrical signal.

3. The image processing device according to claim 1 , wherein the data analyzer statistically analyzes the at least one type of data.

4. The image processing device according to claim 1 , wherein the data analyzer is further configured to analyze data to assess at least one factor among distribution, illuminance, brightness, noise, and sharpness of at least one data among the sensing data and the pixel data processed by the processor.

5. The image processing device according to claim 1 , wherein the controller assesses illuminance among factors related to the quality of the data, and controls a device that is selected according to a range of the illuminance.

6. The image processing device according to claim 1 , wherein the processor comprises, as the AI image processing model, a convolutional neural network (CNN) configured to generate a feature map to analyze the sensing data.

7. The image processing device according to claim 6 , wherein the ALU is at least one of a graphics processing unit (GPU), a field-programmable gate array (FPGA), an application specific integrated circuit (ASIC), or a neuromorphic chip.

8. An image processing device comprising:

a processor configured to process sensing data received from an image sensor; and

a controller configured to control the processor,

wherein the processor comprises:

an artificial intelligence (AI) image processing model trained in image processing through learning;

an arithmetic logic unit (ALU) configured to perform a computation involved in image processing using the AI image processing model; and

an image signal processor (ISP) configured to process an image by using an image processing filter,

wherein the controller is configured to analyze the sensing data in order to select the at least one device among the ALU and the ISP based on a quality of the sensing data and control the processor so that at least one device selected among the ALU and the ISP processes the sensing data, and

wherein the controller assesses noise among factors related to the quality of the sensing data, and controls a device that is selected according to a range of the noise.

9. An image processing device comprising:

a processor configured to process sensing data received from an image sensor; and

a controller configured to control the processor,

wherein the processor comprises:

an artificial intelligence (AI) image processing model trained in image processing through learning;

an arithmetic logic unit (ALU) configured to perform a computation involved in image processing using the AI image processing model; and

an image signal processor (ISP) configured to process an image by using an image processing filter,

wherein the controller is configured to analyze the sensing data in order to select the at least one device among the ALU and the ISP based on a quality of the sensing data and control the processor so that at least one device selected among the ALU and the ISP processes the sensing data, and

wherein the controller assesses sharpness among factors related to the quality of the sensing data, and controls a device that is selected according to a range of the sharpness.

10. An image processing method performed by an image processing device, the image processing method comprising:

analyzing sensing data received from an image sensor;

assessing an image quality by using the analysis result;

selecting a processing device to process the sensing data according to the image quality; and

processing an image by using the selected processing device,

wherein in the selecting the processing device, the following equation is used:

1

N

⁢

∑

i

,

j

⁢

I

⁡

(

i

,

j

)

<

τ

A

where N is the total number of pixels in a target area, and I (i, j) is a pixel value at pixel position (i, j), and

wherein an artificial intelligence (AI) image processing model is used in the case in which an image brightness value of the sensing data is less than or equal to a threshold value as shown in the equation above, otherwise a pipeline-based image signal processor (ISP) is used.