Using digital to correct color-biased images

We know that when you get an image, like a snapshot of a photo, the reflected light illuminates the scene. Or, improper development, printing, scanning and other ways, will make the image color phenomenon. Fortunately, Photoshop has a number of ways to get rid of the colors we don't want in the image. You can easily get rid of those extra colors in a digitally adjusted way. In this tutorial, we have demonstrated how to apply adjustment layers, sampling colors, and information panels to correct the color of the image. Using the digital correction method instead of the visual adjustment, we will not be deceived by our eyes or the display, the results will be more coordinated and more reliable.

1. Use Photoshop to open a pair of color deviation, need to correct the image.

The image in this example is in RGB mode. (Figure 00)

Figure 00

2. Select the Color Sampler tool in the Toolbox and set the sample size to 3x3 average or 5x5 average in the tool options.

Note: If you do your color correction based on a single point, and the point is not representative, then you may get unexpected results. (Figure 01)

Figure 01

3. Measure the color value

Use the Color Sampler tool to click on the high light of the image and place the first color sampler. After the placement, if the position is not appropriate, you can move the sampler to the best position. Then, in the same way, place the second color sampler in the dark section of the image. In our case, we set the high light point at the edge of the plate, and the dark part is located in the darkest portion of the noodle sauce. (Figure 02)

Figure 02

4. Study color values

If you don't see the information Panel on your screen, select the window > display information and open the Info panel. In the panel, you will see the color information in the panel labeled # #, #2的颜色取样器, where it is located, #1和 the light and dark tone of the diagram respectively. Because this image is an RGB image, there is an RGB value in the panel. Each point of the red, green, blue values are different, and when the highlight and darkened values are unified, the image is balanced, there will be no bias phenomenon. It should be noted that sometimes the image needs to be biased, such as the setting of a sunset image, so there is no need to adjust the color of the bias. However, in this image, we need to remove the extra red from the images to make the food look more palatable. (Figure 03)

Figure 03

5. Find the lowest value

In the information Panel, observe the RGB value of a sample point. For example, in the case of high light point, the green is lowest in the red, green and blue values of the sampling point. (Figure 04)

Figure 04

6. Increase the adjustment layer, balance the high light

Select Layers > Create new > Adjust layers, and in type, select a color scale. In the channel option, select the color channel with the lowest value you checked out in the previous step. Here, we choose the green channel. The green value displayed in the Information Panel now is 210. Drag the white triangle in the Color Level dialog box to observe the G value in the info panel until it is equal to the highest value in the RGB value of the high light point. Here, the red value is the highest, so we drag the white triangle to 223, so that in the information Panel, the green value is equal to the red value, and the same is 240. (Figure 05)

Figure 05

Correct other channels: Select the remaining unbalanced channel, which is blue. In the way you just did, move the white Triangle of the dialog box until the information panel shows the same as red (Figure 06).

Figure 06

Now, R, G, b three channels are the same values, the high light part of the color is removed. (Figure 07)

Figure 07

7. Balance Dark Tone

Still in the Color Level dialog box, select the color channel with the highest RGB value for the second sampling point (darkened part). We can tell from the information panel that the red value is the highest. Select the red channel and drag the black triangle to the left of the Level dialog box until the lowest value is the same in the information Panel. (Figure 08)

Figure 08

Next, select the remaining unbalanced channel. Adjust the black triangle until the R, G, and B values are the same in the information Panel (Figure 09).

Figure 09

In this way, #2取样点的RGB值统一以后, the darkened part of the color is removed. (Figure 10)

Figure 10

8. Balanced Intermediate Tone

We also need to eliminate the bias that exists in the middle of the image after the light and darkened parts are eliminated. Select the Color Sampler tool again and place a sampling point in the midtones section of the image. In our example, we put the sampling point on the top right of the plate. (Figure 11)

Figure 11

9. Increase the curve adjustment layer

Select Layer > New > Adjust layer to add a curve adjustment layer. The observation information Panel to determine the median value in the RGB value. In our example, the R, G, and B values are 179,169,174, so we use the B value as the standard. With the information Panel, modify the color channel above or below this value. In our example, we need to reduce the red channel curve while raising the green channel curve. The RGB values in the information panel are the same (Figure 12).

Figure 12

In this way, the color of each region of the image is eliminated. (Figure 13)

Figure 13

The following sections, though unrelated to our subject-using digital correction images, are written out to preserve the integrity of the example as a reference for adjusting the image.

First, flatten all layers. If you want to keep the sampling point, just select another tool, the sampling point will disappear from the screen, select the Color Sampler tool again, the sampling point will be displayed again. Select Filter > Twist > Diffuse Light, in the dialog box, the diffusion granularity is set to 0, the luminous volume set to 5, clear the number of 20, the screen immediately become clean up. Then use the Brightness/contrast command to increase the contrast +10 (Figure 14).

Figure 14

To increase our appetite, we need to make the sauce on the noodles more attractive. With the Lasso tool, select the red sauce body, feather the selection, the radius of 3 pixels on it. With the color level tool, drag the white triangle to 210 and deselect. This will make the sauce look fresher. Then select Vanilla, slightly feathered, and add a bit of saturation to the selection to make the vanilla more colorful. (Figure 15)

Figure 15