Generation and elimination of moire in prepress image processing
The occurrence of moiré patterns is usually produced in three stages of image processing.
1. Scan the texture of the height details
For example, when scanning an image on a textured fabric, moiré occurs, which is caused by the scanner itself. The scanner is not sensitive enough to the details of the reading, and the resulting moire is present in the scanned image and can hardly be removed.
2. Incorrect screening angle output
When a multi-color image is output at the wrong screening angle, moiré appears on the image printed on the printing press. In addition, if the screen angle does not use the optimal value, it will inevitably appear moire.
3. Scan pre-screen images
When scanning an image that has already been printed, the continuous tone information on the photo or transparencies cannot be scanned. Instead, it is a halftone dot. The scanner partially recognizes the space between the dots, but cannot copy it as it is or convert it to continuous tone. As a result, the image contains moire. This situation can use several techniques to reduce or eliminate moire.
Removal of moire
Based on the degree of damage to the overall image quality, several methods for eliminating moiré are introduced in order.
First, reduce the moire when scanning halftone images
When scanning black and white mesh tone images, the following steps usually eliminate the space between dots.
1. Scan the image at twice the normal resolution minus 10 pixels. For example, if a halftone image is typically scanned at a resolution of 300 dpi, it is now scanned at 590 dpi.
2. In Photoshop, select Filter\Blur\GaussianBlur (Filter\Fuzzy\Gaussian Blur). This filter has great damage to the image, so only lower values ​​should be used, and pixels with a radius of 0.7 to 1.2 should be sufficient.
3. Select Image\ImageSize.
4. In the Image Size dialog box, select the "ConstrainProportions" and "ResampleImage" options, you can change the resolution without affecting the width and height, and then set the interpolation to "Bicubic" (Secondary cube).
5. Enter 300 in the Resolution field.
6. Click "OK" to halve the resolution and remove 3/4 of the image pixels. Usually this will also remove most of the pattern.
7. Evaluate the image. If the dot patterns are reduced enough, continue with other editing tasks. If not, proceed to the next step.
8. Evaluate the image again. If there are still a small amount of textures, apply Filter\Noise\Despeckle (filter\varicolor\speckle).
Second, use the rotation to eliminate the moire
When scanning pre-screened color images, moiré not only appears in one channel (black) but may appear in one or more color channels. Over-adjusting the entire image will have an irreversible effect on the information in the channel that did not initially contain the pattern.
Instead of placing the image at right angles to the scanning platform, it is placed at some other angle. This can force the scanner to distort the more obvious component colors, such as blue and black, and blur the dot shape edges during interpolation to create a more continuous tone effect when the image is rotated in Photoshop. It is usually scanned at 15°, 30° or 45°.
After scanning the image, there are two ways to rotate the scanned image. One is to select Image\RotateCanvas\Arbitrary (image\rotate the canvas\arbitrary angle), enter the angle at which the image is to be rotated, and then cut off excess canvases generated after rotating the entire image; and second, crop the image while rotating.
Third, use a fuzzy method to eliminate the moire
The basic steps:
1. Scan the image at a resolution twice the normal resolution minus 10 pixels. In this way, when applying the first round of fuzzy filters, the image contains more information than the standard scan, causing less damage to important details. Also, when the resolution of the blurred image is reduced later, more conflicting dot structures may be eliminated.
2. Blur the image. When you open the scan in Photoshop, use a Gaussian blur filter to slightly distort the dot shape. Generally close to 0.7 pixels can do very well.
3. Verify the color channel. In many cases, moiré only affects one or two channels. Click on the color channel in the Channels panel to check it. After finding the problem color channel, use the above method to apply a Gaussian blur or despeckle filter again to find out and average the abnormal dot values.
4. Reduce the resolution. After blurring the image, reduce the image resolution to the desired value. Select Image\Image Size to select the Constrain Proportions and Resample Image options and enter the new value in the Resolution field. When Photoshop interpolates at the new resolution, each pixel value must be recalculated, which further removes the number of moire. If there is still a slight pattern, check the color channel again and make a Gaussian blur with a smaller value (0.3 to 0.5 pixels).
5. Sharpen the image. When applying a false light mask filter to a de-netting filter, a balance must be struck between emphasizing the details of the image and exaggerating the moiré that cannot easily be removed. When previewing the effect of the filter, set the number value to the beginning of the texture. A previous value.