Image quality enhancement method and device, and electronic device

Abstract

The application provides a picture quality enhancement method and device, equipment and storage medium, which can be applied to various scenes such as cloud technology, artificial intelligence, intelligent transportation, auxiliary driving, video and the like. The method comprises the following steps: a cloud server determines picture quality enhancement information corresponding to a terminal device; if the picture quality enhancement information of the terminal device indicates that a preset picture quality enhancement condition is met, the cloud server skips a picture quality enhancement operation, encodes a target video which has not been subjected to the picture quality enhancement, and obtains a first code stream; and if the picture quality enhancement information of the terminal device indicates that the preset picture quality enhancement condition is not met, the cloud server performs a complete picture quality enhancement operation on the target video, encodes the target video after the picture quality enhancement, and obtains a second code stream. That is, according to the picture quality enhancement information of the terminal device, the cloud server determines whether the picture quality enhancement operation is performed by the terminal device or the cloud server, so that the computing capacity of a decoding end can be fully utilized, the computing resources of an encoding end can be saved, and the encoding efficiency is improved.

Claims

1. A method for enhancing image quality, characterized in that, Applied to cloud servers, including: The process involves obtaining the video type of the target video. If the video type of the target video is a preset type video, then determining the image quality enhancement information and region of interest (ROI) detection information of the corresponding terminal device. The preset type video includes cloud gaming video or vehicle networking video. Determining the image quality enhancement information of the corresponding terminal device includes: sending a first test stream to the terminal device and receiving first information returned by the terminal device. The first information indicates whether the terminal device has an image quality enhancement function, and the first time consumed when performing image quality enhancement on the ROI of the first test video corresponding to the first test stream if the image quality enhancement function is present. Determining the ROI detection information of the corresponding terminal device includes: sending a second test stream to the terminal device and receiving second information returned by the terminal device. The second information indicates whether the terminal device has an ROI detection function, and the second time consumed when performing ROI detection on the second test video corresponding to the second test stream if the ROI detection function is present. If the image quality enhancement information of the terminal device indicates that the preset image quality enhancement conditions are met, and the region of interest detection information of the terminal device indicates that the preset region of interest detection conditions are met, then the cloud server skips the image quality enhancement operation and the region of interest detection operation, and encodes the target video that has not been enhanced to obtain the first bitstream. The preset image quality enhancement conditions include the first time being less than or equal to the preset image quality enhancement time threshold, and the preset region of interest detection conditions include the second time being less than or equal to the preset region of interest detection time threshold. If the image quality enhancement information of the terminal device indicates that the preset image quality enhancement conditions are met, and the region of interest detection information of the terminal device indicates that the preset region of interest detection conditions are not met, then the cloud server performs region of interest detection on the target video, obtains the region of interest information of the target video, sends the region of interest information to the terminal device, skips the image quality enhancement operation, encodes the target video that has not been enhanced, and obtains the first bitstream. If the image quality enhancement information of the terminal device does not meet the preset image quality enhancement conditions, the cloud server performs region of interest detection on the target video to obtain the region of interest of the target video, enhances the image quality of the region of interest, and encodes the enhanced target video to obtain a second bitstream.

2. The method according to claim 1, characterized in that, The image quality enhancement information of the terminal device includes whether the terminal device has an image quality enhancement function, and the image quality enhancement computing capability of the terminal device when it has the image quality enhancement function.

3. The method according to claim 2, characterized in that, The image quality enhancement conditions include preset image quality enhancement calculation capabilities, and the method further includes: If the terminal device has the image enhancement function, and the image enhancement computing capability corresponding to the terminal device meets the preset image enhancement computing capability, then it is determined that the image enhancement information of the terminal device meets the image enhancement condition. If the terminal device does not have the image enhancement function, or if the terminal device has the image enhancement function but the image enhancement calculation capability of the terminal device does not meet the preset image enhancement calculation capability, then it is determined that the image enhancement information of the terminal device does not meet the image enhancement conditions.

4. The method according to claim 3, characterized in that, If the image quality enhancement computing capability of the terminal device includes the first time consumed by the terminal device when enhancing the image quality of the first test video, the preset image quality enhancement computing capability includes the preset image quality enhancement consumption time, and the first test video is the video obtained by the terminal device after decoding the first test bitstream; If the terminal device has the image enhancement function, and the image enhancement computing capability corresponding to the terminal device meets the preset image enhancement computing capability, then it is determined that the image enhancement information of the terminal device meets the image enhancement condition, including: If the terminal device has the image quality enhancement function, and the first time is less than or equal to the preset image quality enhancement time, then it is determined that the image quality enhancement information of the terminal device meets the image quality enhancement condition. If the terminal device has the image enhancement function, and the image enhancement calculation capability of the terminal device does not meet the preset image enhancement calculation capability, then it is determined that the image enhancement information of the terminal device does not meet the image enhancement conditions, including: If the terminal device has the image quality enhancement function and the first time is greater than the preset image quality enhancement time, then it is determined that the image quality enhancement information of the terminal device does not meet the image quality enhancement conditions.

5. The method according to claim 4, characterized in that, The first time includes: the time consumed by the terminal device when performing image quality enhancement on the entire image of each image in the first test video, or the time consumed by the terminal device when performing image quality enhancement on a partial image area of ​​each image in the first test video.

6. The method according to claim 5, characterized in that, The first time includes at least one of the following: the average duration of a single frame, the maximum duration of a single frame, and the variance when the terminal device enhances the image quality of the first test video.

7. The method according to claim 1, characterized in that, The method further includes: If the region of interest detection information of the terminal device does not meet the region of interest detection conditions, then it is detected whether the terminal device has the function of receiving externally input image region information; If the terminal device has the function of receiving externally input image region information, then the target video is subjected to region of interest detection to obtain the region of interest information of the target video, and the region of interest information is sent to the terminal device. The image quality enhancement operation is skipped, and the target video without image quality enhancement is encoded to obtain the first bitstream. If the terminal device does not have the function of receiving externally input image region information, then the target video is subjected to region of interest detection to obtain the region of interest of the target video, and the image quality of the region of interest is enhanced. The image quality enhanced target video is then encoded to obtain the second bitstream.

8. The method according to claim 1, characterized in that, The region of interest (ROI) detection information of the terminal device includes whether the terminal device has a ROI detection function, and the ROI detection capability of the terminal device when it has the ROI detection function.

9. The method according to claim 8, characterized in that, The region of interest detection conditions include preset region of interest detection computing capabilities, and the method further includes: If the terminal device has the region of interest detection function, and the region of interest detection calculation capability corresponding to the terminal device meets the preset region of interest detection calculation capability, then it is determined that the region of interest detection information of the terminal device meets the region of interest detection condition. If the terminal device does not have the region of interest detection function, or if the terminal device has the region of interest detection function but the region of interest detection calculation capability of the terminal device does not meet the preset region of interest detection calculation capability, then it is determined that the region of interest detection information of the terminal device does not meet the region of interest detection conditions.

10. The method according to claim 9, characterized in that, If the region of interest detection calculation capability of the terminal device includes the second time consumed by the terminal device when performing region of interest detection on the second test video, the preset region of interest detection calculation capability includes the preset region of interest detection consumption time, and the second test video is the video obtained by the terminal device after decoding the second test bitstream; If the terminal device has the region of interest detection function, and the region of interest detection calculation capability corresponding to the terminal device meets the preset region of interest detection calculation capability, then determining that the region of interest detection information of the terminal device meets the preset region of interest detection condition includes: If the terminal device has the region of interest detection function, and the second time is less than or equal to the preset region of interest detection time, then it is determined that the region of interest detection information of the terminal device meets the region of interest detection condition. If the terminal device has the region of interest detection function, and the region of interest detection calculation capability of the terminal device does not meet the preset region of interest detection calculation capability, then it is determined that the region of interest detection information of the terminal device does not meet the region of interest detection conditions, including: If the terminal device has the region of interest detection function and the second time is greater than the preset region of interest detection time, then it is determined that the region of interest detection information of the terminal device does not meet the region of interest detection conditions.

11. The method according to claim 10, characterized in that, The second time includes at least one of the following: the average duration of a single frame, the maximum duration of a single frame, and the variance when the terminal device performs region of interest detection on the second test video.

12. A method for enhancing image quality, characterized in that, Applied to terminal devices, including: The method further includes: decoding the acquired bitstream to obtain a target video; if the image quality enhancement information of the terminal device indicates that the image quality enhancement conditions are met, and the region of interest detection information of the terminal device indicates that the region of interest detection conditions are met or not met, then the bitstream is a first bitstream obtained by the cloud server encoding the target video without image quality enhancement; if the image quality enhancement information of the terminal device indicates that the image quality enhancement conditions are not met, then the bitstream is a second bitstream obtained by the cloud server performing region of interest detection on the target video, obtaining the region of interest of the target video, enhancing the image quality of the region of interest, and encoding the enhanced target video; the method also includes: receiving a first test bitstream sent by the cloud server and decoding the first test bitstream to obtain a first test video; when it is determined that the terminal device has image quality enhancement function, calculating the first time consumed when enhancing the image quality of the first test video, and sending first information to the cloud server. The first information is used to indicate whether the terminal device has the image quality enhancement function, and to indicate the first time consumed by the terminal device to enhance the image quality of the first test video when it has the image quality enhancement function; and to receive the second test stream sent by the cloud server and decode the second test stream to obtain the second test video; when it is determined that the terminal device has the region of interest detection function, to count the second time consumed when performing region of interest detection on the second test video, and to send the second information to the cloud server, the second information being used to indicate whether the terminal device has the region of interest detection function, and to indicate the second time consumed by the terminal device to perform region of interest detection on the second test video corresponding to the second test stream when it has the region of interest detection function, the preset image quality enhancement condition including the first time being less than or equal to a preset image quality enhancement time threshold, the preset region of interest detection condition including the second time being less than or equal to a preset region of interest detection time threshold; If the image quality enhancement information of the terminal device indicates that the preset image quality enhancement conditions are met, and the region of interest detection information of the terminal device indicates that the preset region of interest detection conditions are met, then the decoded target video is subjected to region of interest detection to obtain the region of interest of the target video, and the region of interest is subjected to image quality enhancement to obtain the target video with enhanced image quality. If the image quality enhancement information of the terminal device meets the preset image quality enhancement conditions, and the region of interest detection information of the terminal device does not meet the region of interest detection conditions, then the region of interest information of the target video sent by the cloud server is received, and the image quality of the region of interest of the target video is enhanced according to the region of interest information to obtain the enhanced target video. If the image quality enhancement information of the terminal device does not meet the preset image quality enhancement conditions, then the decoded target video will be used as the image quality enhanced target video.

13. The method according to claim 12, characterized in that, The region of interest information of the target video sent by the cloud server includes: If the terminal device has the function of receiving externally input image region information, it receives the region of interest information of the target video sent by the cloud server.

14. The method according to claim 13, characterized in that, If the image quality enhancement information of the terminal device meets the preset image quality enhancement conditions, the region of interest detection information of the terminal device does not meet the region of interest detection conditions, and the terminal device does not have the function of receiving externally input image region information, the method further includes: The entire image of the target video is enhanced to obtain the enhanced target video.

15. An electronic device, characterized in that, include: A processor and a memory, the memory for storing a computer program, the processor for calling and running the computer program stored in the memory to perform the method of any one of claims 1 to 11 or 12 to 14.

Images