Video-based high-image quality image extraction processing method and apparatus, and mobile terminal
AssigneeHuizhou TCL Cloud Internet Corporation Technology Co., Ltd.
InventorNan Xu
Life of this patent
Estimating the term…
Abstract
Disclosed are a video-based high-image-quality image extraction processing method and apparatus, and a mobile terminal. The method comprises: acquiring a video file from which a picture needs to be extracted; selecting a designated segment in the video file for frame-by-frame playback; when a shutter instruction is detected, according to the result of frame-by-frame playback, extracting a key frame required to generate a picture from within the selected designated segment, and extracting specified frames before and after the key frame; and performing multi-frame synthesis on the extracted key frame and the specified frames before and after the key frame, and obtaining a target static image.
Claims16
A video-based high image-quality image extraction processing method, comprising:
- obtaining a video file required to extract a picture;
- selecting a designated segment from the video file for frame-by-frame playback;
- in response to detection of a shutter instruction, extracting a key frame required to generate the picture from the selected designated segment based on a result for the frame-by-frame playback, and extracting one or more front designated frames before the key frame and one or more rear designated frames after the key frame;
- and
- performing multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames to obtain a target static picture;
- wherein the performing of the multi-frame synthesis comprises:
- performing multi-frame synthesis on the key frame as well as the one or more front designated frames and the one or more rear designated frames based on multi-frame noise reduction processing to obtain the target static picture,
- wherein the performing of multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames based on multi-frame noise reduction processing comprises:
- determining an area in which noise points overlap between target consecutive frames, wherein the target consecutive frames are consecutive frames among all of the key frame and one or more front designated frames and the one or more rear designated frames;
- performing color guess processing and interpolation processing on the area to perform noise reduction;
- and
- performing multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames based on a result for the noise reduction to obtain the target static picture.
A mobile terminal, comprising:
- a memory;
- a processor;
- and
- a video-based high image-quality image extraction processing program stored on the memory and operable on the processor,
- wherein the video-based high image-quality image extraction processing program, when executed by the processor, implements a video-based high image-quality image extraction processing method, comprising:
- obtaining a video file required to extract a picture;
- selecting a designated segment from the video file for frame-by-frame playback;
- in response to detection of a shutter instruction, extracting a key frame required to generate the picture from the selected designated segment based on a result for the frame-by-frame playback, and extracting one or more front designated frames before the key frame and one or more rear designated frames after the key frame;
- and
- performing multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames to obtain a target static picture;
- wherein the performing of the multi-frame synthesis comprises:
- performing multi-frame synthesis on the key frame as well as the one or more front designated frames and the one or more rear designated frames based on multi-frame noise reduction processing to obtain the target static picture;
- wherein the performing of multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames based on multi-frame noise reduction processing comprises:
- determining an area in which noise points overlap between target consecutive frames, wherein the target consecutive frames are consecutive frames among all of the key frame and one or more front designated frames and the one or more rear designated frames;
- performing color guess processing and interpolation processing on the area to perform noise reduction;
- and
- performing multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames based on a result for the noise reduction to obtain the target static picture.
A non-transitory computer-readable storage medium having stored thereon a video-based high image-quality image extraction processing program, wherein the video-based high image-quality image extraction processing program, when executed by a processor, implements a video-based high image-quality image extraction processing method, comprising:
- obtaining a video file required to extract a picture;
- selecting a designated segment from the video file for frame-by-frame playback;
- in response to detection of a shutter instruction, extracting a key frame required to generate the picture from the selected designated segment based on a result for the frame-by-frame playback, and extracting one or more front designated frames before the key frame and one or more rear designated frames after the key frame;
- and performing multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames to obtain a target static pictures wherein the performing of the multi-frame synthesis comprises:
- performing multi-frame synthesis on the key frame as well as the one or more front designated frames and the one or more rear designated frames based on multi-frame noise reduction processing to obtain the target static picture;
- wherein the performing of multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames based on multi-frame noise reduction processing comprises:
- determining an area in which noise points overlap between target consecutive frames, wherein the target consecutive frames are consecutive frames among all of the key frame and one or more front designated frames and the one or more rear designated frames;
- performing color guess processing and interpolation processing on the area to perform noise reduction;
- and performing multi-frame synthesis on the key frame as well as the front designated frames and the rear designated frames based on a result for the noise reduction to obtain the target static picture.
Dependencies are inferred from the claim text.