Part 3. : Fundamentals of fundamentals
"Handling images on a computer"

1: Subject 2: Lens 3: Filter 4: Imaging Device 5: Unformatted data

Fig. 1 : Converting an image into a signal (RAW data). An image formed by a lens (2) is sensed
by an imaging device (4) with a filter (3). Finally, RAW data,unformatted data (5),is obtained.

A CCD converts an image into a signal. The signal is converted into RAW data by AD-conversion.
Conventional cameras, digital cameras, and camcorders form an optical image on a plane. The formed image is converted into an electrical signal by an imaging device such as a CCD that is mounted in a digital camera. A CCD is a collection of tiny light-sensitive points (photosites), which convert light into an electrical signal. The brighter light is, the greater a signal becomes.
Each light-sensitive point of a CCD is also called a pixel. The more pixels increase, the finer an image becomes.
In Nikon COOPIX 885/880, you can choose the number of recording pixels to be used.

This means that the size of an image depends on recording pixels (recording size). Roughly speaking, the recording size in digital photography may correspond to a frame size in conventional photography.
By the way, just look at a screen surface of a television set, a computer display, or a LCD display of a digital camera with a magnifier. You can see many tiny points that have one of three primary colors; red (R), green (G), and blue (B).
As described in detail in 3.3.1., the human eye sees an image by combining points composed of three primary colors.
Meanwhile, an imaging device has a filter that composed of many tiny filters that have one of three primary colors. Light that passes through each tiny color filter generates a signal of each color.
There are two types of filters. One is a RBG color system (additive color system). Another is a YMC color system (subtractive color system) in which complementary colors, yellow (Y), magenta (M), and cyan (Y) are used.

1: Various setting values for a camera 2: Image Processing 3: Image of a large file size

Fig. 2 : An image signal is converted into image data (TIFF data). Raw data is converted into TIFF data, reading setting values for a camera (1). An image of a large file size (3) is output.

An imaging device outputs an image signal. The image signal is converted into digital data by an A-D converter. This digital data is called RAW data.
RAW data, roughly speaking, may correspond to a latent image in conventional photography. That is, we cannot see an image from RAW data like we cannot do so from a latent image.
So, RAW data must be converted to a visible type of data. Although we have learned to set various settings in Part 1. and 2., these setting information is needed when RAW data is converted. An arithmetic unit (it may correspond to a film processor in conventional feelnikon) of a digital camera converts RAW data into image data, referring to the various setting information. If Exposure Compensation or White Balance has not been set correctly, desired tones or colors cannot be obtained.
In the case of Nikon D1 series, the camera can take and record a picture using RAW data. But in order to see an image on your computer, you need to use a special program such as Nikon Capture 2 that converts RAW data into an image (it may correspond to film developer in conventional feelnikon).

1: Setting values in image quality 2: Compression 3: Image of a small file size

Fig. 3 : Image data is compressed into JPEG data. TIFF data is compressed to produce JPEG data, reading setting values in imge quality (1). An image of a small file size (3) is output.

Thus, RAW data is converted into visible type data. Typical visible type data is TIFF data. For example, COOLPIX 885/880 uses TIFF data when image quality is set to [HI] (non-compression). Disadvantage of TIFF data is to need a large capacity to include image data (a large file size). So when you use TIFF data, the total number of shooting frames fairly decreases and it takes a considerably long time to store an image into a storage medium due to the large amount of image data.
Digital cameras generally compress image data to decrease the amount of data, thereby resulting in good usability. Many digital cameras, including COOLPIX 885/880, adopt the JPEG format that decreases the amount of data maintaining image quality as high as possible.
In the JPEG format, compression ratio is selectable. When image quality is set on a digital camera, compression ratio is simultaneously set. In the case of COOLPIX 885/880, compression ratio is as follows:

Such compression in which original data cannot be restored is called non-reversible compression. Although there are many compression formats other than the JPEG format, the JPEG format seems to be most rational and pervasive.
Now we have finished the JPEG format and are ready to advance to the next step.

If you don't much care about practicality, it is better to compress data at lower compression ratio. When image data is compressed at low compression ratio, it is easy to create an image at desired quality. On the other hand, when image data is compressed at high compression ratio, it is difficult to do so.
But if you consider practicality to be important, in the case of COOPIX 885/880/995, it is better to use FUUL (SUXGA) as upper limit size and [FINE] as upper limit image quality. For me, it is inconvenient to use the camera at [HI] in image quality mode. Because the total number of frames is small and the time required to store an image is long.

If you use images to send e-mail or create a Web page (Home page) but not to print pictures, it is sufficient to use XGA (1,024 x 768 pixels) image size and [NORMAL] image quality for COOLPIX 995/885/880 even if image quality deteriorates.
Next, I introduce how to compress an image file for a Web page with maintaining image quality (it is called sampling down). Small file size makes transmitting time very short. Now let's open Adobe Photoshop.

Using the "Image Size" command for sampling down

Using the "JPEG Options" for sampling down

Procedures:

1. Set a magnification to 100% using the "Navigator" palette, or double-click the zoom tool icon. Then, an on-screen image becomes actual size.
2. Choose Image > Image Size to open the "Image Size" dialog box.
3. Make sure that the "Constrain Proportions" checkbox and the "Resample Image" checkbox are selected, and the "Resolution" box is set to 72 dpi (dots per inch, 1 inch is about 25.4 mm). The 72 dpi resolution is suitable for viewing an image on a monitor.
4. Enter appropriate values into the "Width" and "Height" boxes of "Pixel Dimensions" and click the "OK" button to change an image size to a desired size.
   It is better that an image size is smaller than VGA size (640 x 480 pixels) for use in a Web page. When an image is larger than VGA size, you may need to scroll the screen to see the whole image.
5. Choose File > Save As to use the "JPEG Options" command.
6. Set the "Quality" box to "JPEG Medium" or "JPEG Low" and save the image.

Next, I explain the adjustment of resolution when an image is printed by a home-use printer.
Unlike the use in a Web page, it is better that image quality should precede file size for purpose of printing pictures.

Using the "Image Size" command for printing

Using the "JPEG Options" for printing

Procedures :

1. Deselect "Resample Image", then "Width", "Height", and "Resolution" are connected by a chain.
2. Enter 200 - 360 dpi into "Resolution". At this time, a print size is changed, but leave it as is.
   The larger a value of "Resolution" is, the finer print quality becomes. The finer print quality becomes, the smaller print size becomes. It is not reasonable to set the resolution value higher than that of printer performance.
3. Then, change a print size to a desired size.
   Select "Resample Image" and then enter values into "Print Size". When a print size is large, pixels are interpolated by software. But this does not mean that more details are added to an image.
4. Choose File > Save As to open the "JPEG Options" dialog box. Select "Maximum" and save the image.

Beginners tend to capture too many subjects in one picture. So, when you look at such a picture, it looks very confused. And you wonder what is an aim of this picture.
One good way to take a better picture is to avoid capturing too many subjects in one frame. So, it is said conventionally that photography is art to subtract extra factors from the frame.
But executing the subtraction in a frame is really difficult. In particular, when you are a photographer just starting taking pictures, you probably cannot afford to do so.
But you don't need to worry about it because you can crop an image using Adobe Photoshop.
By cropping an image, you can eliminate an extra subjects in an image and show only the main subject.
When you succeed in cropping an image, you have a well simplified image. And, as you are getting accustomed to the cropping, your framing technique on shooting will also be getting better.

The crop tool (left upper) and the Options palette (right upper)

By using the crop tool, an unnecessary guy (?) in the left area has been eliminated.

Procedures :

1. Click and hold down the "Rectangular Marquee" tool in the toolbox, then the hidden tools appear. And select the "Crop" tool in the hidden tools.
2. When you want to crop an image keeping an aspect ratio constant, select the "Fixed Image Size" check box and click the "Foreground" button.
3. And drag the Crop tool diagonally to select a desired area, then the desired area is surrounded by a border.
4. When you want to move the selected area, click and drag the area inside the border.
5. When you want to resize the selected area, drag a corner (a control handle) of the border.
6. When you want to rotate the selected area, move the pointer outside the selected area, then pointer becomes double headed arrow. And rotate the selected area using the double headed arrow.
7. When the desired area is correctly selected, press the Enter key (Mac: Return key), then cropping is executed. When you want to interrupt cropping, press the Esc key (Mac: esc key).

The "Rubber Stamp" tool is a tool that samples part of an image and paints with a copy of the sampled area. This tool lets you easily retouch acne and freckles on the skin or flaws on the surface of a product.
But to avoid leaving the trail of retouching, it is necessary to use various option functions or to practice the retouching technique to some degree.

"Rubber Stamp" tool (left upper), "Options palette" (right upper), and "Brushes palette" (right upper-middle)

An image retouched by the "Rubber Stamp" tool

Procedures:

1. When you want to retouch details, magnify an area you want to retouch.
   
2. Select the Rubber Stamp tool.
3. Select soft-edged brush (blurred outline) in the Brush palette.
4. Hold down the "Alt" key (Mac: "Option" key). Click the area to be sampled, at this time, the black triangle of the Rubber Stamp tool turns white temporarily.
   It is better to sample an area near to a retouching area.
5. Place the Rubber Stamp tool over the retouching area and click the mouse button. At this time, the sampled area is indicated by a cross-hair pointer.
6. If you drag the Rubber Stamp tool, retouching is continuously executed. If you click once, only one area is retouched.
7. To avoid leaving the retail of retouching, repeat to retouch around the area.

Adobe Photoshop has various tools. Now I introduce some retouching techniques similar to conventional retouching techniques. Select a tool suitable for a retouching type, select a retouching level or a retouching area in the options palette or the brush palette, and click and drag the retouching area.
Operations are simple but it is necessary to practice operations to some degree to use the tools effectively.

Blur, Sharpen, and Smudge tools
When you hold down the Blur tool, the Sharpen tool appears.

The position of the Blur tool. Jeans under watch are blurred.

 

1. Blur tool
   The Blur tool blurs too sharp area or too high contrast area of an image. If you repeat to click, degree of retouching is increased.
   If you select a brush size in the Brushes palette and opacity in the Options palette, you can adjust degree of retouching.
   The position of the Sharpen tool and its effect
   The dial of a watch is sharpened.

 

2. Sharpen tool
   

   The Sharpen tool has reverse effect to the Blur tool. The Sharpen tool can sharpen a blurred area to some degree.

   The position of the Smudge tool and its effect
   Jeans are smudged.

 

3. Smudge tool
   Click and drag an area to be retouched using the Smudge tool, then you have a smudged effect as colors are mixed. By using the Options palette, you can use other colors.

Toning tools (Dodge, Burn, and Sponge tools)
When you hold down the Dodge tools (default setting) of the toolbox, the Burn and Sponge tools appear.
If you have an experience to work in the dark room to process conventional photographs, these three tools are very familiar tools for you. Also use the Options palette and the Brushes palette effectively.

The position of the Dodge tool. The upper part of the image is a little brightened.

1. Dodge tool
   The Dodge tool can brighten part of an image. By using it, shadow part of an image can be made clear and a white object is more whitened. When you set the Pressure slider, you can adjust the intensity of the toning effect. When you use the Brushes palette, you can select brush size.

The position of the Burn tool and its effect. The upper part of the image is a little darkened.

2. Burn tool
   The Burn tool can darken part of an image. By using it, you can make clear the highlight part of an image and more darken shadow part.
   The Option and Brushes palettes are used in the same way as the Dodge tools.

The position of the Sponge tool and its effect. The saturation of the burn of Manjuu (Japanese cake) is increased.

3. Sponge tool
   The Sponge tool can change saturation. By using the Sponge tool, you can adjust saturation of an image. When you adjust the Pressure slider in the Options palette, saturation is increased or decreased.

First, I'll begin with the structure of the human eye in order to explain about color.

Fig. 4 Simplified structure of the human eye. 1. cornea, 2. crystalline lens, 3. glass body, 4. retina, 5. cone, 6. rod

I explained earlier that a digital image consists of three primary colors. This is not limited to the digital image. Analog images, including conventional photography and printing, also basically consist of three primary colors. Why ?

Because human eyes perceives three primary colors and human brain forms an image from them.
Fig. 4 is a cross sectional view of the human eye which you sometimes see somewhere. The cornea and crystalline lens corresponds to a lens of a camera. The retina corresponds to film or an imaging device in a conventional camera or a digital camera.
The retina has two kinds of cells which perceive light.
One is rod cells and another is cone cells. The rod cells are sensitive to light, so they work well even in dark conditions. But the rod cells cannot perceive colors. The reason we doesn't perceive much color at night is that the rod cells cannot perceive color and perceive only light.
The cone cells are sensitive to colors and work at brightness more than a certain level. The cone cells, interestingly, consist of three types of cells that perceive three primary colors respectively, i.e., Red, Green, and Blue. The human brain forms various colors based upon light intensity that three-type cone cells receive.
That is, the combination of three primary colors, Red, Green, and Blue, can form every color that the human can perceives. This is the principle of the color system that a computer display, a digital camera, and conventional photography adopt. These three primary colors are called additive primary colors. In the additive primary color system, all colors are created by adding three additive primary colors together at some ratio.

Fig. 5 :
A ring of three primary colors (Color wheel). Learning a position of each color is useful in understanding color mechanism.

There is another type of three primary colors system, i.e., subtractive primary color system. The subtractive primary colors are colors obtained by subtracting the additive primary colors from white light. They are Cyan (C), subtracting Red from white light, Magenta (M), subtracting Green from white light, and Yellow (Y), subtracting Blue from white light.
Subtractive three primary colors are often used for printing and painting. Referring to Fig. 5 makes it easy to understand color mechanism.

Supplement:
Relation of complementary colors is given by the following formula.
R (red) = - C (cyan)
G (green) = - M (magenta)
R (Blue) = - Y (yellow)

Let a state of no color be zero, then
R+C=0 G+M=0 B+Y=0 R+G+B=0

Thus, colors can be computed.

"Variations command"
Each number corresponds to order of the procedure.

As described previously, use of additive primary colors and subtractive primary colors enable to produce a wide variety of colors. Strictly speaking, the range of colors that the human eye can perceive is much wider than that produced by the additive primary colors or subtractive primary colors that are used in photography or printing.
But at least in photography and printing, colors are created from the three primary colors. That is, colors in photography can be created at will by using three primary colors.

Using the "Variations command" of Adobe Photoshop, which is based upon three primary colors, even a beginner can correct colors adequately. The Variations command displays multiple versions of the same image so you can easily identify the desired color adjustment. You can also adjust brightness and saturation of an image.

The Variations command opens through Image > Adjust > Variations.

Procedures:

1. The upper left image is an original image, and the image next to the original image is the current image that is being corrected.
   If adjusting work becomes confused, click the original image, then the Variations command restores a starting state.
2. Select one of "Shadows", "Midtones", and "Highlights" to set tones for color balancing. First, select Midtones, then try Shadows or Highlights.
3. You can change an adjusting width using the slider.
   One scale of the slider corresponds to changing two times (or one-second). For a beginner, it is better to change greatly.
4. Seven images below the above-described two images are variations in color balance. These images are arranged like the ring of three primary colors (color wheel). When you click an image close to your desired image, the selected image is placed in the center of the ring which is the current image, and new variations are displayed.
5. Three images at the right of the seven images are variations in brightness. When you click an image close to your desired one, the current image turns the selected one.
6. When you select "Saturation", variations in color balance disappear and three images appear which are variations in saturation. Click an image close to your desired one.
7. When the current image becomes your desired one, click "OK."

You may sometimes become confused during complicated adjustment. At such a time, it is better to compare the current image with the original image or real objects around you.

And to a beginner, I recommend changing greatly. One good way is to change greatly first and then adjust gradually. It is a short path to obtain a good result.

Variations command can easily adjust color balance, brightness, and saturation. But you sometimes cannot get a desired image. In this case, try to use other various tools such as described below. The description allows you to get desired results. At first, you may have unsatisfactory results, but continue to use the tools referring to this article. As you repeat operations, you are getting accustomed to the tools and you'll have good results at last. My motto is "all's well that ends well".

Levels command
A profile shaped like a mountain is a histogram of an image.

1. Levels command and Auto Levels command
   Choose Image > Adjust > Levels, or Image > Adjust > Auto Levels.

Auto Levels command
You can automatically adjust tones and colors of your image only by selecting "Auto Levels."

The "Auto Levels" command automatically performs the "Levels" command. The feature of the Auto Levels is similar to the feature of the "Auto" in a camera. The Auto Levels work fairly well. If an image cannot be corrected satisfactorily by the Auto Levels command, then use the Levels command.

The Levels command adjusts tones and contrast of an image.
A graph shaped like a mountain is called a histogram that displays the pixel values that are plotted along the horizontal axis (brightness).

If a mountain locates in the left area of the graph, the graph means that dark tones are superior in the image.
On the contrary, if a mountain locates in right area of the graph, the graph means that light tones are superior in the image.
Understanding a histogram is useful for adjusting tones and contrast of an image.
When you adjust tones and contrast, use the "Eyedropper" tool and the slider. And by changing "Channel", you can adjust tones and contrast for each of three primary colors.

The Curves command
Let's understand the meaning of the Curves.

2. Curves command
   Choose Image > Adjust > Curves. The Curve, briefly speaking, is a function that adjusts the characteristic of a digital image that corresponds to the film characteristic in conventional feelnikon.

3. Color Balance command, Brightness/Contrast command, and Hue/Saturation command
   Choose Image > Adjust > Color Balance (Brightness/Contrast, Hue/Saturation)
   Although adjustment in the Variations command is not continuous, it is possible to adjust an image continuously in the above commands. When you become accustomed to the Variations command, then use these commands.

   By the way, Adobe Photoshop Elements, which is the latest version of Adobe Photoshop 5.0 LE, don't have the Curves command. Instead, Adobe Photoshop 5.0 LE can darken a highlight area by using the "Adjust Backlighting" command and lighten a shadow area by using the "Fill Flash" command.

Color Balance command
You can continuously adjust the mixture of colors in the image with the Color Balance command.

Brightness/Contrast command
You can fast adjust brightness and contrast of an image.

Hue/Saturation command
If you are familiar with the Munsell system of color, the Hue/Saturation command may be easier to use than the Color Balance command.

Now I have completed to describe the basic handling of a digital image on computer. I wonder if you become a little confused. But I believe it is important to practice basic techniques even if the practicing takes a time to some degree. Until now, I have been studying digital image processing on computer. I believe there is no royal road to learning the digital image processing.
You must be careful saving adjusted image data. Don't use Save because it eliminates an original image. Instead, use Save As to save an original image.