Search

CN-122023140-A - Image processing method, device and program product

CN122023140ACN 122023140 ACN122023140 ACN 122023140ACN-122023140-A

Abstract

The disclosure provides an image processing method, an image processing device and a program product, relates to the technical field of image processing, and is used for improving the effect of image processing. The method comprises the steps of obtaining an image pair set, wherein the image pair set comprises at least one image pair, each image pair comprises a first image and a second image, the image definition of the second image is larger than that of the first image, the image alignment processing is carried out on each image pair in the image pair set to obtain aligned images in each image pair, the image fusion processing is carried out on the aligned images in each image pair and the first image to obtain a processed image pair set, the processed image pair set comprises at least one image pair, and the matching degree between the second image and the first image in the processed image pair is larger than that between the second image and the first image in the image pair before processing.

Inventors

  • REN KE
  • XIE CHANGJIANG

Assignees

  • 中兴通讯股份有限公司

Dates

Publication Date
20260512
Application Date
20241112

Claims (11)

  1. 1. An image processing method, the method comprising: Acquiring a set of image pairs, the set of image pairs comprising at least one image pair, each image pair comprising a first image and a second image, the second image having an image sharpness greater than the image sharpness of the first image; Respectively carrying out image alignment processing on each image pair in the image pair set to obtain aligned images of each image pair; and performing image fusion processing on the aligned images and the first images in each image pair to obtain a processed image pair set, wherein the processed image pair set comprises at least one processed image pair, and the matching degree between the second image and the first image in the processed image pair is greater than that between the second image and the first image in the image pair before processing.
  2. 2. The method according to claim 1, wherein performing image alignment processing on each image pair in the image pair set to obtain an aligned image in each image pair, includes: Aiming at each image pair in the image pair set, aligning a first image to the first image based on the image characteristics of the first image, the image characteristics of a second image and an optical flow estimation technology to obtain an aligned image, wherein the alignment precision of the aligned image and the first image is higher than that of the second image and the first image, and the matching degree of the aligned image and the first image is higher than that of the second image and the first image.
  3. 3. The method of claim 2, wherein the image features include at least one of field of view, color, brightness.
  4. 4. The method of claim 2, wherein aligning the second image to the first image based on image features of the first image, image features of the second image, and optical flow estimation techniques in the image pair, resulting in an aligned image, comprises: Estimating a high-precision optical flow field from the first image to the second image based on the image features of the first image, the image features of the second image and an optical flow estimation technology in the image pair; and according to the high-precision optical flow field from the first image to the second image, reversely mapping each pixel point in the second image to the first image so as to map each pixel point in the second image to the corresponding pixel point position in the first image, thereby obtaining the alignment image.
  5. 5. The method of claim 4, wherein estimating the high-precision optical flow field from the first image to the second image based on the image features of the first image, the image features of the second image, and the optical flow estimation technique in the image pair comprises: Extracting features of the first image and the second image respectively to obtain a plurality of image feature layers of the first image and the second image, wherein different image feature layers correspond to different image resolutions; An initial optical flow estimation step, namely obtaining an initial optical flow field from the first image to the second image based on an image feature layer with the minimum image resolution and an optical flow estimation technology; An up-sampling step, namely carrying out interpolation up-sampling treatment on the initial optical flow field to obtain an up-sampling optical flow field, wherein the image resolution of the up-sampling optical flow field is larger than that of the initial optical flow field; A feature fusion step of carrying out fusion processing on the image feature layer corresponding to the image resolution of the up-sampling optical flow field and the up-sampling optical flow field to obtain a fused optical flow field; Iterative optimization, namely taking the fused optical flow field as a new initial optical flow field; And repeatedly executing the up-sampling step, the feature fusion step and the iterative optimization step until the image resolution of the new initial optical flow field reaches the image resolution of the first image, and determining the new initial optical flow field as a high-precision optical flow field from the first image to the second image.
  6. 6. The method according to claim 1, comprising low-frequency luminance information of an image, wherein said performing an image fusion process on said aligned image and said first image in said respective image pairs to obtain a set of processed image pairs comprises: separating the low frequency luminance information of the first image and the aligned image for each of the image pairs; generating a uniform background image according to the first image from which the low-frequency brightness information is removed; combining the first image from which the low-frequency brightness information is removed and the aligned image from which the low-frequency brightness information is removed with the background image respectively to obtain a combined first image and a combined aligned image, and determining the combined first image as the processed first image; carrying out fusion processing on the chromaticity information of the combined first image and the brightness information of the combined aligned image to obtain a processed second image; and obtaining the processed image pair according to the processed first image and the processed second image.
  7. 7. The method of claim 2, wherein aligning the second image to the first image based on image features of the first image, image features of the second image, and optical flow estimation techniques in the image pair, resulting in an aligned image, comprises: based on the view field information of the first image and the view field information of the second image, clipping the second image to obtain a clipped second image, wherein the view field of the clipped second image is consistent with the view field of the first image; performing color and brightness migration processing on the cut second image based on the color and brightness information of the first image, or performing brightness migration processing on the cut second image based on the brightness information of the first image, so as to obtain a migrated second image; and aligning the second image after the migration processing to the first image based on the image characteristics of the first image in the image pair, the image characteristics of the second image and the optical flow estimation technology to obtain an aligned image, wherein the matching degree of the aligned image and the first image is higher than that of the second image and the first image.
  8. 8. The method of claim 1, wherein the set of image pairs comprises image pairs under different acquisition scenes, the acquisition scenes being associated with at least one of: lens type, lens focal length information, illumination environment, jitter degree, shooting angle, depth of field range and object motion state.
  9. 9. The method according to claim 1, wherein the method further comprises: training the initial model based on the processed image pair set to obtain an image processing model, wherein the image processing model is used for deblurring of the image.
  10. 10. An electronic device comprising a memory and a processor, the memory and the processor being coupled, the memory being for storing instructions executable by the processor, when executing the instructions, performing the method of any one of claims 1 to 9.
  11. 11. A computer program product, characterized in that the computer program product comprises a computer program which, when run on a computer, causes the computer to perform the method according to any one of claims 1 to 9.

Description

Image processing method, device and program product Technical Field The present disclosure relates to the field of image processing technologies, and in particular, to an image processing method, apparatus, and program product. Background With the development of science and technology, mobile devices are becoming more popular, and have become a major tool for people to record life. Thus, the user's requirements for the camera function of the mobile device have increased. However, in the shooting process, the situations of insufficient illumination, poor depth of field control, lens distortion, aberration and the like may exist, so that the problems of blurred image edges, poor overall definition and the like are caused, and the visual experience of a user is affected. At present, by using an image processing method based on an image processing model, fuzzy-clear image pairs acquired under a large number of scenes can be used for learning the characteristic mapping of image deblurring, so that the image definition is improved. But the deblurring level of the image processing method based on the image processing model may be low, subject to the acquired blurred-sharp image pair. Disclosure of Invention The present disclosure provides an image processing method, apparatus, and program product for at least enhancing the effect of image processing. In order to achieve the aim, the first aspect provides an image processing method, which comprises the steps of obtaining an image pair set, wherein the image pair set comprises at least one image pair, each image pair comprises a first image and a second image, and the image definition of the second image is larger than that of the first image; respectively carrying out image alignment processing on each image pair in the image pair set to obtain aligned images in each image pair; and performing image fusion processing on the aligned images and the first images in each image pair to obtain a processed image pair set, wherein the processed image pair set comprises at least one processed image pair, and the matching degree between the second image and the first image in the processed image pair is greater than that between the second image and the first image in the image pair before processing. The image processing device comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring an image pair set, the image pair set comprises at least one image pair, each image pair comprises a first image and a second image, the image definition of the second image is larger than that of the first image, the processing module is used for respectively carrying out image alignment processing on each image pair in the image pair set to obtain aligned images in each image pair, the processing module is also used for carrying out image fusion processing on the aligned images in each image pair and the first image to obtain a processed image pair set, the processed image pair set comprises at least one processed image pair, and the matching degree between the second image and the first image in the processed image pair is larger than that between the second image and the first image in the image pair before processing. In a third aspect, there is provided an electronic device comprising a processor and a memory, the memory storing instructions executable by the processor, the processor being configured to, when executed, cause the communication device to implement any of the methods as provided in the first aspect above. In a fourth aspect, there is provided a computer readable storage medium storing computer instructions that, when run on a computer, cause the computer to perform any of the methods provided in the first aspect. In a fifth aspect, there is provided a computer program product comprising computer instructions which, when run on a computer, cause the computer to perform any of the methods provided in the first aspect. Based on the technical scheme provided by the disclosure, the image pair is processed, so that the obtained processed image can keep the authenticity and naturality, high-quality image output is realized, and the image processing effect is improved. In addition, the matching degree between the clear image (second image) and the blurred image (first image) in the image pair is improved, and the image processing effect of the image processing model can be improved by training the image processing model through the processed image pair. Drawings The accompanying drawings are included to provide a further understanding of the disclosed embodiments and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain, without limitation, the disclosed embodiments. FIG. 1 is a schematic diagram of an image processing system according to an embodiment of the present disclosure; Fig. 2 is a schematic flow cha