Image reflection elimination method based on panoramic camera

Abstract

The image reflection elimination method based on panoramic photography involves the field of image processing technology, and solves the corresponding problems existing in the reflection elimination method used in the prior art. It consists of four steps: data preprocessing, rough reflection alignment, fine reflection alignment, and background image recovery. , the method proposed in the present invention uses panoramic photography technology to introduce the content information of the reflection scene through the panoramic image, which greatly reduces the ill-posedness of reflection elimination, and makes the effect of reflection elimination greatly surpass the method of only using a single mixed image, avoiding the confusion due to Issues with degraded reflection removal results caused by background scenes and reflective scene content. A reflection alignment method is proposed to convert the reflection scene in the panoramic image into the corresponding glass reflection image, and the designed background restoration network restores a high-quality background image guided by the glass reflection image. Using a single panoramic image with glass reflection interference areas greatly reduces the difficulty of data acquisition.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image reflection elimination method based on panoramic photography. Background technique [0002] With the development of computer technology, computer computing power is gradually strengthened, machine learning and deep learning technology are advancing rapidly, and computer vision-related technologies are gradually applied to various scenarios, such as face detection of mobile phone cameras, retouching and beautifying pictures, and taking pictures at night. Pedestrian detection and road recognition in human driving, face recognition for mobile payment and station identity detection, or simultaneous positioning and mapping tasks for robots, etc. With the advent of the era of big data and intelligence, more and more application scenarios need the support of computational vision technology. Massive video and image data need to be processed urgently, which highlights th...

AI Technical Summary

Problems solved by technology

However, both methods have their own disadvantages: 1) It is an ill-posed problem to use a single color image for reflection elimination. Confuse the background layer with the reflection layer, resulting in the failure of reflection elimination; 2) Using multiple images for reflection elimination can achieve better reflection elimination effect than the single-image method, but it needs to take multiple images, and the data acquisition process is cumbersome. It has high requirements for shooting equipment and photographers

The shooting process is complicated, data acquisition is difficult, and the requirements for photographers and shooting equipment are very high

[0030] 3. Existing technology 3 uses the image sequence with reflection acquired by an ordinary camera to perform reflection elimination, and the camera needs to be in a stable moving state when acquiring the image sequence. When the camera is not stable enough or moves too fast, the relative motion of the background layer and the reflection layer The amplitude will increase, resulting in a significant decrease in the effect of reflection cancellation

Therefore, this technology has high requirements for photographers and shooting equipment, and cannot be a stable and reliable reflection cancellation solution

Images