Scanning transmission image restoration method and device and storage medium
A transmission image and recovery method technology, applied in the field of image processing, can solve the problems of difficulty in data set acquisition, difficulty in effectively removing scanning transmission image information, difficulty in generalization, etc., to achieve good separation effect, high generalization ability, and improved performance force effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0057] In view of the problem that the text content on the back side will be transmitted to the front page during the scanning process of thinner paper printed matter or written matter (for example: ordinary paper books, teaching materials, test papers), thereby affecting the normal reading of readers, the embodiment of the present invention provides A scanning transmission image restoration method is proposed, see figure 1 , the method includes the following steps:
[0058] 101: Construct a preset network composed of a generator network and a discriminator network;
[0059] Wherein, the preset network includes: a generator network and a discriminator network, and the generator network includes: a first generator G d and the second generator G h ; The discriminator network includes: the first discriminator D d and the second discriminator D h , set the generator network and the discriminator network to the confrontational learning mode, the generator network and the discri...
Embodiment 2
[0094] The scheme in embodiment 1 is further introduced below in conjunction with specific calculation formulas, accompanying drawings, and examples, see the following description for details:
[0095] 201: The problem of scanning transmission image restoration can be regarded as a conversion problem from image to image, namely: the problem of converting a scanning transmission image to a non-scanning transmission image;
[0096] In the process of training the model, a large number of scanned transmission image data sets are required as the data basis. However, there is no directly usable scanned transmission image dataset in reality. Therefore, the first step is to construct a dataset of scanned transmission images. In order to ensure good generalization performance of subsequent models, the types and styles of scanned transmission images should be as many as possible. Collect different styles of text materials and make scanning transmission images. In order to ensure the g...
Embodiment 3
[0117]Below in conjunction with concrete evaluation standard, accompanying drawing and table, the scheme in embodiment 1 and 2 is carried out feasibility verification, see the following description for details:
[0118] In order to objectively evaluate the restoration effect of the transmission image, the peak signal-to-noise ratio (PSNR, Peak Signal-to-NoiseRatio) and structural similarity (SSIM, Structural SIMilarity) are used as the evaluation criteria for the restored image quality, and the given size is m×n pixels Image I and image K, PSNR and SSIM are defined as follows:
[0119] (1)
[0120] in, is the mean square error, I(i, j) and K(i, j) are the pixel values of image I and image K at position (i, j) respectively.
[0121] (2)
[0122] Among them, μ I is the mean value of I, μ K is the mean of K, is the variance of I, is the variance of K, σ IK is the covariance of I and K, c 1 =(k 1 L) 2 , c 2 =(k 2 L) 2 It is two constants to avoid dividing by...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


