Image processing method and system

1. An image processing method, comprising:

determining whether each pixel of an input image belongs to a target object of the input image using a segmentation algorithm, wherein the target object is composed of pixels belonging to the target object;

determining an enhanced image of the target object, wherein the enhanced image of the target object is an image in which each pixel of the target object is displayed in an enhanced manner; and

determining a positioning image of the target object by applying an integral image algorithm to the enhanced image of the target object;

wherein determining whether each pixel of the input image belongs to the target object of the input image using the segmentation algorithm comprises:

inputting the input image into a deep convolutional neural network to perform a pixel-level feature extraction on the input image; and

determining whether the pixel belongs to the target object based on a feature value of the pixel;

wherein determining the enhanced image of the target object comprises:

performing a feature extraction on the input image to determine a pixel feature graph;

performing a feature extraction on the input image to determine a context feature graph;

determining context relation information of each pixel based on the pixel feature graph and the context feature graph; and

determining the enhanced image of the target object according to the context relation information of each pixel of the input image;

wherein each pixel of the input image carries weight information, and determining the enhanced image of the target object according to the context relation information of each pixel of the input image comprises:

changing weight information of the pixels belonging to the target object, so that the target object is displayed in the enhanced manner.

2. The method according to claim 1 , wherein determining the positioning image of the target object by applying the integral image algorithm to the enhanced image of the target object comprises:

determining an integral image according to the enhanced image of the target object; and

determining the positioning image of the target object by using the integral image.

3. The method according to claim 2 , wherein determining the integral image according to the enhanced image of the target object comprises:

applying a scale factor to the enhanced image of the target object.

4. The method according to claim 1 , wherein the method further comprises:

calculating a loss rate between the enhanced image of the target object and the input image based on a loss function; and

feeding back the calculated loss rate to the segmentation algorithm to update the segmentation algorithm.

5. The method according to claim 1 , wherein the segmentation algorithm is implemented by a deep convolutional neural network HRNet18.

6. An image processing system, comprising:

a memory storing computer-executable instructions; and

a processor coupled to the memory,

wherein the computer-executable instructions, when executed by the processor, cause the system to perform the following operations:

determining whether each pixel of an input image belongs to a target object of the input image using a segmentation algorithm, wherein the target object is composed of pixels belonging to the target object;

determining an enhanced image of the target object, wherein the enhanced image of the target object is an image in which each pixel of the target object is displayed in an enhanced manner; and

determining a positioning image of the target object by applying an integral image algorithm to the enhanced image of the target object;

wherein determining whether each pixel of the input image belongs to the target object of the input image using the segmentation algorithm comprises:

inputting the input image into a deep convolutional neural network to perform a pixel-level feature extraction on the input image; and

determining whether the pixel belongs to the target object based on a feature value of the pixel;

wherein determining the enhanced image of the target object comprises:

performing a feature extraction on the input image to determine a pixel feature graph;

performing a feature extraction on the input image to determine a context feature graph;

determining context relation information of each pixel based on the pixel feature graph and the context feature graph; and

determining the enhanced image of the target object according to the context relation information of each pixel of the input image;

wherein each pixel of the input image carries weight information, and determining the enhanced image of the target object according to the context relation information of each pixel of the input image comprises:

changing weight information of the pixels belonging to the target object, so that the target object is displayed in the enhanced manner.

7. The system according to claim 6 , wherein determining the positioning image of the target object by applying the integral image algorithm to the enhanced image of the target object comprises:

determining an integral image according to the enhanced image of the target object; and

determining the positioning image of the target object by using the integral image.

8. The system according to claim 7 , wherein determining the integral image according to the enhanced image of the target object comprises:

applying a scale factor to the enhanced image of the target object.

9. The system according to claim 6 , wherein the computer-executable instructions, when executed by the processor, further cause the system to perform the following operations:

calculating a loss rate between the enhanced image of the target object and the input image based on a loss function; and

feeding back the calculated loss rate to the segmentation algorithm to update the segmentation algorithm.

10. The system according to claim 6 , wherein the segmentation algorithm is implemented by a deep convolutional neural network HRNet18.