High-level image processing technology breakthrough

Some people may think that using a 16-bit (which can represent 216 = 65536 levels) amount of data to describe the tone level from pure black to pure white is a little overkill, because there is almost no visually between the 16-bit image and the 8-bit image. What is the difference, and most output devices cannot copy more than 8 bits of color. That's true, but if you've ever edited an image in 16-bit mode, you'll be amazed at its merits. It can provide 256 times the image accuracy of an 8-bit image.

Each pixel in an 8-bit grayscale scan image may have 256 luminances (0-255); in an 8-bit color scan image, each pixel has 3 color channels (8 bits per channel, a total of 24 bits) Corresponding to any one channel, 256 luminances (0-255) may also appear for each pixel.

Common misunderstandings

Many people will hold this view: "The bit depth is equal to the dynamic range." This is not true. Dynamic range refers to the difference in density between the brightest and darkest areas of a subject (digital camera), film (slide or negative), or reflection (print). The bit depth refers to the number of steps that fall within the entire density range. The accuracy of the image is increased in the 16-bit file, making the transition between the image tonality and the gradation smoother. Of course, its cost is relatively high.

Saying goodbye to color jumps

If there are not enough step steps in an area of ​​the image, there will be color jumping, or the contrast between the tone is enlarged, the visual effect is very bad, and the quality of the image is greatly reduced. The 16-bit file adds thousands of steps to the image. There is no need to worry about jumping between colors. The details of the hierarchy are more abundant. Maybe you have to say that the device can only output 8-bit files, ie 16-bit data editing Is the document inconvenient? Don't worry, more and more equipment manufacturers have begun to pay attention to the problem of bit depth. Every prepress creator should learn from now on and learn how to use the 16-bit model to get more levels of tonality.

By looking at the scan results of different manuscripts (transmissions and reflections), let us know how to use dynamic range and bit depth to obtain high quality digital images.

Usually, the black area of ​​the transmitted document is relatively deep, which requires the scanner to have a relatively large dynamic range. If the Kodachrome or Velvia slides and the negative film are viewed under the lamp table at the same time, the range of transmissive film gradation is very large. At this time, only the scanner with a larger dynamic range can record richer image details.

In contrast, the maximum density of the negative film should be softer, and more benefits can be obtained from high-level scans. Since the negative film is not critical to the scanner's dynamic range, the tone curve can be applied to adjust the image level so that the contrast between the highlight area and the dark tone area is large enough. However, the stretching of the tone will cause the 256 steps in the 8-bit image to be insufficient, resulting in some undesirable phenomena, such as the increase in the contrast between jumps or tones, or giving a false impression. feel.

The situation of the reflection draft is relatively simple, and only the scanner with a relatively dynamic range and a relatively small number of digits can complete the scanning process, and it is easy to be realized by the printing machine.

What is a rough sweep?

If you do not perform a high scan under the control of the scanning software, this scan method is called rough scan. When you open a rough scan file in Photoshop, you will be very surprised and even shocked. Negative scans are very bright; positive drafts look very dark. no need to worry.

All scanning software provides editing functions, such as curve, level and color correction, allowing the operator to modify the image tone and color in the scan preview state. In addition to these commonalities, there are significant differences between the various scanning softwares in the process of acquiring image hue. They are generally divided into 5 groups:

1. Old-fashioned, inexpensive scanners can only capture 8-bit data in one pixel. In this kind of equipment, you cannot enter the world of 16-bit data. (Note: Do not confuse this file with the 8-bit file captured by the new scanner. The new device's scanning software is an 8-bit image obtained by downsampling.)

2. Some medium dynamic range scanners can capture 10, 12 or 14-bit images, but only 8-bit files can be output. In this case, you can use the scanning software to get the 8-bit file that is most suitable for Photoshop. Due to the old-fashioned model, this medium dynamic range scanner cannot produce high-order original files.

3. The third type of scanner can generate high-level files, and can introduce extra bit depth into Photoshop. To do this, you need to turn off all editing features in the scanning software and generate a high-level original file. (The rough scans look very dark (transmittal), or very light (negative)). Note that you can only export an 8-bit file to Photoshop if any of the adjustment features are included in the scan software preview.

4. Another type of scanner can capture raw data and attach a feature description file to the final image file. If you use a scanning software that contains a characterization file of the photographic material, you can output an image file that is better than the original image's color and contrast.

5. The last type of scanner captures high-level images and can edit them in a high position before exporting to Photoshop. The Imacon Flextite and Umax PowerLook scanners are of this type. Although adjusting the tone before scanning can save a lot of time, it is important to keep in mind that you can only handle high-level information in the scanner's software. You can also use the Curves and Levels commands in Photoshop to work with rough scans.

Get rid of 16-bit links

If the Macintosh menu is grayed out, it cannot be applied; if you click the Quick Mask icon and the "Cannot use Quick Mask function" prompt dialog box appears, you have entered the 16-bit file kingdom .

Although there are significant restrictions on editing 16-bit images in Photoshop 6.0, there are some important tools for handling 16-bit files, including Unsharp Mask, Gaussian Blur, Dust and Scratches filters, and High Pass.

Some people think that in the 6.0 version missed a high edit command, that is, the Fade command in the edit menu. This manual feature can adjust the individual functions of the lower-level files, from false light masks, curves, to color balance effects. In future versions of Photoshop, it is likely to increase its ability to handle high-level images and wait patiently.

Although there are many tools that cannot be applied to the processing of high-level images, the application of only a few tools can also achieve good results. Let's take a look at the two fastest and simplest tools.

Select from 8-bit images by clicking and dragging

Copy the 16-bit image (Image->Duplicate) to bring up the Copy Image dialog box and name it; then turn it into an 8-bit image (Image->Mode->8bits/channel). The 8-bit image and the 16-bit image are displayed on the screen at the same time, and the 8-bit image is treated as a template, after which the selected image is clicked and dragged onto the 16-bit image.

Save selection

If you need to create multiple mask selections on a 16-bit image, the most effective way is to save the selection (secec->Save Select) like saving an 8-bit template image. As long as the 8-bit image is open, the selected content can be loaded into a 16-bit image (selec->Load Selection).

Although this method is slightly more benzene than the click-and-drag method, it can save a lot of time when several areas need to be selected during the production process.

There will certainly be more bit depth captures and corresponding output devices in the future, so it is now necessary to familiarize yourself with the rich tone-level images. Learning to use 16-bit mode, you will be competitive in digital imaging.