If multiple devices are used in the image system, must the profile be used for color images? If so, where do you get the properties file?
Color management using ICC profiles is very versatile and can be used with scanners, digital cameras, printers and displays. The resulting or produced profiles and application software can support color management. In general, digital camera profiles are difficult to make unless it is used in a studio with a fixed lighting environment.
Many devices use sRGB as the standard color space. When the data is converted from sRGB or sRGB, manufacturers have done a reasonable job to ensure that they produce "acceptable" high-quality colors. sRGB is designed as a standard color space for use by users and networks in the same way as ICC PCS, but it depends on the interconversions the vendor establishes in their device drivers, independent of the user-supplied profile. Therefore, if you confirm that you have set up an sRGB workflow, you can get high-quality images from these transformations. Due to the limitations of many conditions, digital camera sRGB conversion is very difficult for manufacturers, because the digital camera driver must compensate for various lighting conditions, but this conversion can still be quickly recognized by many people. Of course, if you don't like the conversion provided by the manufacturer, there are only two choices. If the results obtained are all consistent "inaccurate", you can define an editing bar for use in all images in the image editing program, or use a profile.
Is there a tool for editing profile files?
Most software packages that allow the creation of profiles are provided with editing tools. Some software packages are flexible and convenient to use. Such as Kodak Colorflow is the most common software. Most profile creation software provides some editing features. When you want to edit a property file, you must have some color knowledge in advance, and you need to know some of the editing skills and experience. When editing feature files, be sure to carefully and carefully avoid mistakes.
How to change the lookup value of the characteristic file? Which profile editor (software) supports this method?
At present, no software packages have been found to directly change the lookup table (LUT) value of the profile. If anything, from the point of view of image quality, doing so is dangerous and difficult to define, unless you really know what to do, and given the size of many tables, do it slowly and find that the mistake is corrected immediately. The best way is to use the feature file package to indirectly change the value by correcting the properties of the profile.
I recently bought some feature file software for making printer feature files. I want to put some software on a website, but I know that this is not allowed due to software license restrictions. The ICC profile specification is an open standard. In this case, can the software vendor use it? Is it possible to edit a generated profile with a different profile editing tool? Can I create a feature file myself so I'm not limited by the original feature file?
In the use of the characteristic file, ICC has no formal description, it is decided by the software manufacturer. However, because the software vendor effectively holds the copyright of the feature file (specified in a tag on the profile), they have a software license that prohibits the public release of the profile. One of the reasons is that if such feature files can be freely exchanged, their software sales will be greatly reduced, and their economic benefits will be affected. Also, from a technical point of view, publishing the characteristics of many devices is dangerous. For example, the printer's profile is only valid for its substrate and special inks. For the above reasons, there are very few device manufacturers who publish the feature files of their devices.
Any profile is made using proprietary software. All ICC definitions are tag-based, some tags are mandatory, some tags are optional, and how the data is defined in these tags. The contents of the table are vendor-specific and each table uses a different algorithm, which gives the vendor some copyright.
If you edit a feature file with another vendor's software, the situation is complicated. Editing software uses its own copyright label. If the original feature file generated the object code, strictly speaking, editing the feature files of different vendors in this way is not legal, and it is actually impossible for such object code to appear.
After making the ICC profile, why evaluate the profile?
After the ICC profile is created, it is better to evaluate the characteristics because it is influenced by many factors, such as not paying attention to changing the parameters, or changing the parameters with time, etc. It cannot be completely guaranteed. file. The evaluation profile generally refers to the evaluation of its accuracy and its ability to simulate a particular color. The simple method is to measure with a measuring device. If each color block is measured, the software will send it to the raw data of the display, compare it with the measured data, and then display the calculated color difference ΔE value. It can also be calculated mathematically. The ΔE value should be controlled within a certain range.
In order to improve the quality of printed matter, China has set clear requirements for the color difference in the 1987 “GB7705-87 in offset prints” and “Toppan prints GB7707-87” issued by the Ministry of Light Industry and issued by the National Bureau of Standards. The color-difference requirements of the same batch of color decoration prints are: ΔE of a general product is 5.00 to 6.00, and ΔE of a fine product is 4.00 to 5.00.
When the ΔE value we get exceeds a certain range, we should recreate the ICC profile.
How can I use the CIELAB values generated by the input properties file to compare with the original values?
This requires a CIELAB to CIELAB profile (assuming the color management software has this capability) to use as an output profile. Then use Phtoshop's "Convert to profile", use the absolute colorimetric method and no black dot compensation to combine the input profile with it. At this time as a CIELAB image displayed in Phtoshop, these values can be obtained. However, please note that only colorimetric matching methods can be evaluated, and sensory color matching methods cannot be evaluated.
What is color difference? How to calculate the color difference of two colors?
Chromatic aberration is the difference in color perception between two colors, which reflects the closeness of the two colors. It can be calculated mathematically and its value is expressed by ΔE. CIE L*a*b* values are often used to evaluate the degree of difference between prints and proofs. L*, a*, b* in the CIELAB color space can be calculated using the tristimulus values according to the method specified in ISO 13655. Calculate CIELAB based on two sets of color coordinate values (L*(1), a*(1), b*(1)) and (L*(2), a*(2), b*(2)) Color difference ΔE*ab. among them:
Lightness difference: ΔL*=L*(1)-L*(2), where (1) and (2) represent two samples.
Chromaticity difference: Δa*=a*(1)-a*(2)
Δb*=b*(1)-b*(2)
Total color difference: ΔE=SQRT{[L*(1)-L*(2)]2+[a*(1)-a*(2)]2+[b*(1)-b*(2)] 2}=SQRT[(ΔL*)2+(Δa*)2+(Δb*)2]
How to calculate the ΔE of a feature file gamut and another gamut?
Measuring the difference between two color gamuts is not easy. However, the CIELAB color difference value ΔE*ab can be calculated by referring to the method for calculating the color difference described above.
ΔE = SQRT{L*(1)-L*(2)]2+[a*(1)-a*(2)]2+[b*(1)-b*(2)]2}
Here (1) and (2) are used to represent two samples.
For comparison with a profile, an IT8.7/1 color scale can be used, processed with the same input profile, and compared with different output profiles. The final color was measured for each print, and the difference in color was calculated using the above formula. Therefore, it is possible to find the color difference between the excellent block A1, the color difference between A2, and the difference between other color blocks. Then take the average of the entire map as the total color difference between the measurements.
To preview the color gamut, you can use the measured data to make a graph of a* and b* data. However, if you only use primaries and secondary colors to draw the color gamut, it is easy to see the differences in the outstanding domains. For most devices, this is a reasonable approximation. However, it should be noted that the color gamut is three-dimensional, and such rendering only shows two-dimensional ones. To get a three-dimensional representation, you also need to plot the C* value of each hue relative to L*.
C*=SQRT{[a*(1)-a*(2)]2+[b*(1)-b*(2)]2}
Note: This still does not measure the difference between outstanding domains. It is also required to calculate the volume of the odd-shaped entity from these measurements, which is quite complicated. It can do this: Calculate the volume of the 12 shapes defined by white and black and divide the entity into 12 small parts. The black, primary and secondary colors together form the physical shape and add them together.
What methods do you use to test the performance of the feature file software?
We already know that one method is to analyze the ΔE value, and the problem of analyzing the ΔE value is an image of the color gamut. For colorimetry, this is perfectly possible. But you can't use the sensation method, because if you are a portrait photographer, you must pay attention to the correct skin tone and you may ignore other colors. If you have a good gamut mapping algorithm, the CIELAB value can be calculated "ideally" after the color gamut is mapped, used to test the image, and then the ΔE value is calculated. In fact, very few people do this.
The most straightforward way to evaluate the software performance of the production profile is to process a real image and observe it, rank it, compare practices and evaluate it, or compare the print with its original to determine if the profile is acceptable .
