Sorry, this is a really old page that is going to be rewritten. Thank you for your patience!

The Digital Workflow

by Patrick Young

Created: June 13, 2004
Last Edited: July 10, 2004


The negative is the equivalent of the composer's score, and the print the performance. - Ansel Adams

Introduction

This article is not a detailed examination of the features of Photoshop nor of the techniques described; it is a general overview of these and specific techniques used in my own image post-processing workflow. This article is targeted toward advanced-novice/intermediate level photographers.

Many people go out with their camera, take pictures and think that's pretty much the end of it. The download their pictures into their computers, resize them and upload them somewhere to share with others or make some prints if they have a photo printer. Many don't even do that; these days you can take your pictures and have them printed out immediately, whether digital or good old film.

When they get their pictures back, there can be somewhat of a letdown; the prints just don't look right for some reason. "Why did that picture turn out they way it did--it looked so nice on the LCD?!" The big misconception is that the most important aspect of photography is obtaining the photograph; people take for granted how important image post-processing, or time spent in the darkroom, really is. Post processing is extremely important; as much as taking the picture itself. Ansel Adams spent much of his time perfecting his skill in the darkroom as much as out the field. The time and effort you spend in the darkroom is reflected ultimately in your final product--whether it be a print or a JPEG image.

This is a presentation of the workflow I use to prepare my photos for display. All my post-processing is done on computers, in the "virtual darkroom". I don't have a real darkroom because I don't have space or the money for it, plus in this day and age it is the images you can get using computers can be good, if not better than anything you produce from a traditional darkroom. And it cannot be argued how efficient and powerful modern digital photo processing is; it simply cannot be matched by traditional techniques.

There are a lot of good books and other resources out there; definitely do your research to find what's best for you. Hopefully this will give you a few ideas to get you started.

What You Need To Get Started

The digital darkroom has never been easier or cheaper to assemble. Modern personal computer technology has advanced so far that just about any personal computer will due. Here are the basic things you need to get started:

  • Personal Computer - PC or Apple Macintosh, it doesn't really matter. Most modern personal computers are more than adequate. If possible, try to load up on RAM (Random Access Memory) and disk space, photo imaging is always a RAM hog and can take up a lot of your hard drive space. A good barometer is to have a minimum of 256Mb of RAM. 512Mb would be nice, and 1Gb would be heaven--simply the more the better.
  • Photo imaging software - Obviously THE software to use is Adobe Photoshop. The current version is CS (v8.0) although older versions are adequate, but best to use something at least version 7.0. Photoshop is expensive, so there are alternatives, like Adobe Photoshop Elements (consumer version of Photoshop), JASC Paint Shop Pro, Corel Photo Paint, and a host of others.

In this guide, we'll be using Adobe Photoshop CS using a PowerMac G5 running Apple OS X.

Importing Your Images

The obvious first step is to import your images into computer. If you have a digital camera, this is very easy. If you use film, you'll need to scan your images with a photo scanner or have them sent to a photo lab and have them scan it for you. The import of source material is very important, so let's take a closer look.

Importing from Film

Digital cameras may be taking over, but a lot of us (including myself) still use film. You can do a lot with your images on film, provided you can import them into your computer. Two ways to go about it:

Send it To a Photo Lab

The easiest way is to send them out to a photo lab and have them scan it for you. Scanners can cost a lot, so if you don't have money for one, this is a cheap alternative. Also, the process of scanning film is VERY tedious, so if you don't have the time, this also is the way to go. I don't have experience with this since I scan my images myself at home, but it shouldn't be hard to find photo labs that will do this for you. I know my local Costco will do it, so how hard can it be?

Scan it yourself

Scanner technology has really progressed a great deal; you can find cheap scanners everywhere. If you want to go this route, it's important to get the best scanner you can for the money. Try to get a film scanner, that is a scanner that can scan the negative itself. Scanning a print really is not the way to go; your imported original can never be (and often never is) as good as the original you scan from--so your quality in this case will never exceed that of the print.

There are two very important factors to consider when looking at scanners, optical resolution and dynamic range. Resolution describes the precision by which your scanner is able to scan your film, the higher the better. It is described as DPI (dots per inch). Thus a DPI of 2600 means that for each inch of film, the scanner is able to resolve 2,600 pixels. Only optical resolution matters; ignore digital or interpolated scan DPI's--they really are just a marketing gimmick. A good DPI for example for film scanning would be at least 2600 DPI. Realize that a standard monitor is 72 DPI, these scanned images will be very large indeed!

Dynamic range is also very important. This describes the ability of your scanner to differentiate tonal ranges--in particular in darker, shadow regions. This is often noted as the scanner's D(max) value. A D(max) of 3.6 is quite good for example. The higher the D(max) value, the better the scanner is able to resolve these darker areas--and you will notice the difference readily. In addition, D(max) values, like values on the Richter scale are logarithmic the difference between D(max) 3.6 and D(max) 3.7 is very, very significant. And of course, you have to pay for that extra sensitivity--and it can be quite shocking sometimes. Dynamic range is as important if not more important than DPI.

Good scanners also come with nifty features such as dust spot removal/compensation. No matter how hard you may try, dust is a fact of life an you can count on it sitting on your film not matter how hard you try keep things absolutely clean. And at such high scanning resolutions, dust spots really show up. Some scanners will come with such technology as Digital ICE, which removes the spots for you; not a feature whose value should be underestimated. Some scanners also support multi-scanning, which is the scanners will scan your negative many times and average the pixel values out. This helps reduce image noise at the cost of speed but it is quite effective for darker images.

The scanner I use is the Nikon Super Coolscan LS-2000 which I bought in 1998 has a D(max) of 3.6. It has Digital ICE, multi-scan interpolation and a host of other features. In 1998 it cost me US$1600. Today the newest version of this scanner, the Nikon Super Coolscan V ED costs about $600--with 4000 DPI and D(max) of 4.2. Not too shabby.

Remember color profiles!

No matter which way you import your material from film, it's important to have your images accompanied by color profiles. More on color profiles later; just know that color profiles are absolutely critical to insure accurate color representation. Make sure any scanner you use has a color profile; any place you take your images to have them scan, be sure to check that the images will come with a profile.

Saving Your Scanned Images

Most of the time it's a good idea to save your imported scanned images before you start working on time (in some cases, where you have a separate scanner software, you might have no choice). You should always save your images in a lossless format with embedded color profiles. My favorite format for this purpose is the TIFF format. TIFF allows you to save your images without lossiness, with compression and with embedded color profiles. We'll take a closer look at using TIFF's and color profiles a little later.

Importing from a Digital Camera

It does not get any easier than using a digital camera. Simply plug your camera into your computer and import the image directly; or use a card reader and read the storage media directly. All if not the great majority of digital cameras come with color profiles or use standard color profiles, which makes like a whole lot simpler too. When taking pictures, there are a couple of things to think about:

RAW versus JPEG format

If you are in a rush, need space or aren't concerned that your images need much post-processing exactness, JPEG format will work for you. In fact, many digital cameras only save images as JPEG's. But if quality is of utmost importance, you have the time and patience to post-process your images and you have a lot of space, then RAW images are the way to go. Let's look at some of the differences, the advantages and disadvantages of both:

JPEG Advantages

  • compressed so they use up less space on your digital media; important if you have limited space
  • fast, they are often easier for the digital camera to deal with since they are smaller
  • easier and faster to deal with on your computer

JPEG Disadvantages

  • compression leads to immediate lossiness of image quality
  • all image parameters (white balance, exposure compensation) are immediately applied
  • 8-bit color

RAW Advantages

  • raw information directly from the imaging sensor which can be processed later by computer rather than the camera
  • lossless image (maximum image quality)
  • white balance, exposure compensation, contrast and saturation and other parameters are stored as settings in the RAW file, but are not applied
  • ability to work with 16-bit color

RAW Disadvantages

  • very large, uncompressed data that can be unwieldy
  • proprietary format that almost invariably unrecognized

If you have the time and space, the RAW is the way to go. RAW images are lossless raw data without any special handling applied by the camera. So what? RAW data is unprocessed data which is interpreted by special computer algorithms and programs to create the output image you see. When you use JPEG mode, you are using the digital camera's processor and firmware to process the image for you at the time of shooting. If you use RAW mode, you are actually delaying this process; using programs on your computer to do process the information. Your computer is much more powerful than your digital camera so it can apply much more complex algorithms, and often it is the case that software (Photoshop RAW converter for example) can do a much better job. Plus you have the ability with RAW importing software to make adjustments such as white balance, contrast and saturation to customize the import algorithms. This kind of power cannot be achieved using JPEG's.

Importing JPEG Images

Digital cameras these days are good enough to embed color profile information into the images they generate. Sometimes however, they don't have one, or they don't match the one in current use by Photoshop. In that case you will get the following dialog box:

What do you do? Well if you are lucky, your image already contains a working profile so you can simply convert your image to the working profile (that dialog is slightly different than the one above, that one gives you the opportunity to convert to the working profile).

If you aren't lucky and you don't have a profile, then you are in a bit of a bind. My solution is to use the monitor's profile (in the above example it is called the "SyncMaster") and then have it converted to the working RGB. If you try that and it looks wrong, you might just try blindly assigning the working profile to the image. You should never leave your image unmanaged.

Color profiles will be explored in more detail later in this article.

Importing Using the Adobe Photoshop RAW converter

With Adobe Photoshop CS (and Photoshop 7.0 with the RAW converter plug-in installed) you can load RAW files easily. When you do load a RAW image (on the Photoshop menu, choose File -> Open), you will get a dialog box with several options you can play with.

Using the RAW converter is quite straightforward despite the incredible control you are presented with. The settings choose are immediately reflected in the preview window. When you are ready to import your image, simply click 'OK' and the RAW converter will apply its algorithms to the RAW information to generate an image which open in Photoshop. One of the nice things you will notice is that color profile information (in this case, it is indicated by the color "Space" chooser tool) is handled automatically for you.

For more information about using the RAW Converter, check your Photoshop documentation or look around the Internet, there are plenty of resources detailing this powerful tool.

Color Profiles: What They Are and Why They Are So Important

Color profiles have been mentioned so many times now already; what the heck are they? Simply put, they are the means to achieve color accuracy. Color profiles (or color spaces as they are sometimes referred) are the face of a somewhat overlooked but vitally important component of digital imaging--color management. Color profiles ensure that a print or JPEG image of your hand indeed looks like your hand. Color management can be hard to understand, but it actually is quite simple.

Components, Tonality and Color Gamut

To begin, let's look at the basics. A component is defined as any input or output device that will handle your image during your imaging workflow. Here are some examples:

Digital Workflow Components
Film Workflow Components
Digital Camera Film Negative
Computer Monitor Film Scanner
Photo Printer Computer Monitor
  Photo Printer

The object of color management is to ensure the color accuracy throughout the chain, from the source to the final product. No components are created equal; there are even differences between individual monitors of the same model and design! (Color management is a science in itself and you could go as deep as you want to ensure color accuracy; here we will deal with fairly idealized situations where the assumption that associated color profiles perfectly or near-perfectly match the devices they are associated with)

Each type of device, indeed each available model of a device usually has different tonality and color gamut capabilities. What is color gamut and tonality? Color gamut describes the range of colors that the device is able to output. Tonality describes the range of contrast a device is able to display. For example some devices are not able to output certain colors (such as a photo printer, which can never display a pure red) so it's color gamut is limited. Older monitors for example cannot display certain darker shades of colors; they simply become black regions--this is a limitation of tonal range.

The purpose of color management is to be able to map colors between these devices so that the output of each matches the original despite the limitations of each device. Your original often exists in a color space much larger than the spaces any of your devices are capable of. This is where color profiles (ICC profiles as they are referred) come into play.

Magic of Color Profiles

As your image moves along your workflow, it is accompanied by its color profile. So long as you work in a color profile, you can almost be certain color accuracy will be retained. When you scan an image, it should have a color profile assigned to it, often the profile of the scanner itself. Same with a digital camera, its profile should be assigned or embedded with the image. If not, you should have the profile available and ready to associate with the image.

Indeed film itself has its own color profile; different films have different tonal ranges and color gamuts.

Let's take a look at a common scenario using a film scanning situation:

  • Start with an image you intend to scan, let's say it is on Fuji RVP (Velvia) film.
  • Scan it with your film scanner, the Nikon Super Coolscan LS-2000. After you scan it, it should be assigned the ICC profile for the Nikon scanner. Save the image as a TIFF, or it is automatically imported into Adobe Photoshop.
  • In Photoshop, the scan is converted to a much larger color space for digital imaging work. Let's choose to convert to the much larger color space Adobe RGB (1998).
  • Do your image processing, saving the image in TIFF format, saving the image with the ICC profile Adobe RGB (1998) embedded into it.
  • You want to make a version for the web. As a shortcut, you can use "Save for the Web" and Photoshop will do all the color management for you. Otherwise, convert the image to the ICC profile sRGB and "Save" the image as a JPEG--do not overwrite your original TIFF image. Use the resulting JPEG for your things involving the Web.
  • Now you want to make a print. Select "Print Preview" and under "Color Management" choose the ICC profile for your photo printer (the "Source Space" should be automatically set to Adobe RGB (1998); if not then change it to be so. Make your print.

With digital images, especially those done in RAW, much of the initial color space selection process is done automatically for you. If you consistently follow this workflow, preserving the color space along your workflow, your colors will retain correct color accuracy. Note though, older or limited devices may display your images wrong; color management can only do so much--if the device does not support color management or is severely limited then much of this is awash.

General Guidelines Using Color Profiles

  • Make sure that a color profile accompanies all images you create or scan. If they are not already embedded with the image, then you should have one readily available which Photoshop can use as a guideline for proper conversion. ICC profile files are often distributed with the product (monitor, digital camera, printers); if you can't find it you can most likely download it for from from the manufacturers website. Many profiles already come pre-installed with your operating system or with Adobe Photoshop. Never leave an image unmanaged by color management.
  • When working with Photoshop, always convert your image to a larger color space. This gives you more flexibility while you work since your range of colors is now much larger. Although your monitor may not be able to display the colors (Photoshop actually renders the image in your monitor's color space, so you often cannot see the difference) working in the much larger color space is important for precision. A good large color space to choose (in fact I use it as my default) as a working color space in Photoshop is the Adobe RGB (1998) color space.
  • Don't alter your original by overwriting it. Always use this working version; if you mess up you can always start over by re-importing the original.
  • When saving your work, always save it in a lossless format which preserves your color space at minimum. A good choice is to use the TIFF format. Remember, this is your working version that has maximum detail done in a large working color space.
  • Create special purpose versions of this working version for such things as prints and web images. From your working version, you can easily create versions for the web and for printing--just be sure to convert to color profiles that fit for each. For the web, use sRGB. For printing, be sure to select the destination profile as your printer's profile.

Working on Your Image

All that and now finally to actually working on your image. With all your color profile stuff handled, it's now time to start on touch ups and edits.

There are an infinite number of things you can do to fix up your image:

In this article, we'll look at the first and last items--fixing colors and sharpening. We will look at two very simple techniques to enhance the color of your images, Levels and Curves, then look at the Unsharp Mask tool.

Understanding and Using the Levels Tool

The Levels tool is presented under the "Image" menu, under the "Adjustments" submenu. Choose "Levels" to access the Levels Tool. Open it and you will get something like this:

 
The levels tool corresponding to the image on the right. The levels diagram is a presentation of the image's histogram.

 

The levels tool is a presentation of your image's histogram. There are plenty of resources about histograms, so we won't go into detail about there here; basically a histogram represents the tonal values from black (emptiness of light) to white (100% light). The graph represents to total tonal range available; 0% light is on the left side and 100% light is on the right. The arrows beneath the graph represent the endpoints of the tonal range utilized by the image. The median (or middle) arrow indicates the midpoint or the idealized "medium gray" vertice.

So what does all that mean? The entire graph represents the tonal range available to your image, the area between the arrows indicates the area being utilized by your image. The middle arrow indicates where your images tonal midpoint is. In photographic terms, this is the point your camera estimated was the best place to maximize the tonal range of your resulting image--according to Ansel Adams zone system this is 18% gray or Zone V (5). With the levels tool, you have the ability to modify the tonal range of your image!

Well not quite, you can only modify it on a limited basis of course; you can never exceed the tonal range of the original; only define a subset or variation. But this is still quite a powerful tool; you can make the colors of your images much more vibrant by using this tool. Let's play with the levels tool, first let's modify the utilized tonal range by shrinking it and examine the results:
Shrinking the utilized tonal range, in this case by moving the white (100% light) point makes the image much brighter. Any tone that exists beyond the white point washes out. Increased brightness is due to the midpoint moving back from the tonal range collapse, such that darker tones have become the mid tone. The effect of this modification is increased contrast. This is analogous to using a higher contrast film, overexposing the image to gain detail in shadow areas and sacrificing highlight detail.
In this image we shrink the utilized tonal range by adjusting the black (0% light) end point. This causes the opposite effect as the first modification. We have increased contrast by collapsing the utilized tonal range. Consequently, the midpoint moves into the the highlight areas, making the image darker. Note the lost of detail in darker areas; any tones to the left of the black arrow become black--a loss of shadow detail. This is analogous to using higher contrast film and underexposing to obtain detail in the highlight areas.
In this example. the endpoints are not modified, so the utilized tonal range matches the total available tonal range. However the midpoint has been moved into darker tonal areas. The result is the midpoint is now the darker areas, thus darker tones become the midpoint tones. The affect of this action is to stretch and compress the darker and lighter regions on either side of the midpoint. Thus 50% of the total tonal range reserved for darker areas has now been compressed, while the other 50% has been stretched out for the lighter areas. The effect is obvious; the image appears brighter because amount of total tonal range (about 66%) is now reserved for the light colors. This is called a change in gamma. (If you've ever wondered what gamma was, well this is it...) Notice closely that our utilized tonal range remains the same--that is no highlight or shadow detail has been lost from the original.

Not bad, and we're just getting started! As you can see this is one way to optimize image contrast. By simply shrinking the utilized tonal range, you can increase color contrast and saturation of an image, making colors much more vibrant, the image brighter or darker. Ideally with this technique, you should shrink the utilized tonal range such that the majority of tonal range extends between the two endpoints. If you look at the original histogram, there is a large area in the highlight tonal range that has minimal usage. By shrinking this area, you can maximize contrast without sacrificing a great amount of highlight detail, as exemplified below:

That's basically all there is to leveling! You can an awful lot more; what we've talked about is simply control of the overall tonal range. You also have the ability to control the tonal range of individual channels--Red, Green, and Blue. You can experiment with these by simply using the "Channel" chooser tool to pick which channel you wish to modify. All the same things we've discussed here apply; they simply apply only the specific channel you modify.

Automated Leveling

If you are lazy like me, you want the computer to do this all for you. Photoshop can do that for you too. You can just have do it automatically by pushing the "Auto" button, and you can see immediately what Photoshop thinks is best. Or if you want to assert a bit more control over automation, you can click on the "Options" button (in Photoshop v7.0 and later)

http://www.creativepro.com/story/feature/17164.html

With the "Auto Color Correction Options" dialog, you can tell Photoshop they way you want things done automatically. Let's take a look at the available algorithms:

  • Enhance Monochromatic Contrast - Optimize the overall tonal range (maximize contrast)
  • Enhance Per Channel Contrast - Optimize the tonal range of each color channel (maximize per channel contrast)
  • Find Dark and Light Colors - Optimizes tonal range and color by searching for shadow, highlight and midpoint colors based on settings in the "Target Colors & Clipping" settings (new in Photoshop 7.0)

By playing with these settings, you can save a lot of time and really improve the look of your image. Of course, you'll always have much more control if you did it manually.

Understanding and Using the Curves Tool

Like the Levels tool, the Curves tool is found under the menu item "Image", under the "Adjustments" submenu. Click on "Curves" to bring up the curves tool:

The curves tool modifies the brightness of different tonal regions. When you initially open the curves tool, you will see the above, with a diagonal line going from the bottom left to the upper right corner of the graph. The graph represents tonal range (x-axis) and luminosity (y-axis). The plain diagonal line represents neutral or unmodified luminosity across the entire tonal range.

A good idea is to always use the curves tool after the levels tool. Curves does not affect an image as drastically as the Levels tool does.

Dragging the curve higher makes that region more luminous; conversely dragging it lower makes it less luminous. If you were to drag the curve at a point in the red (or highlight tonal region) higher, that region would get brighter. Drag it down it would get darker. The more vertical the slope of the line, the greater the contrast in the tonal region represented by that area.

  • Red: Highlight tones
  • Blue: Midtones
  • Green: Shadow tones

 

The curves tools allows you to modify the luminosity of different tonal regions of your image.you For example, if your image has dark shadow regions you want to make brighter you could use the curves tool to accomplish this. Let's take a look at how this is done.

 Here is a picture of my friend Carrie. Notice the very strong shadows all around the image. We would like increase the luminosity of the shadow regions because we would like to see more detail in the image and reduce the amount of contrast. Before we do that, let's learn more about the Curves Tool and what it does...

Let's open up the curves tool and start playing around with it.
In this image, we've dragged the curve somewhere in the highlight region higher. Notice how highlights get brighter, to the point where highlight detail is lost. Notice how the curve bends. Photoshop does this to avoid posterization.
Here we do the opposite of the first image. By dragging the highlight area down, we can see that the highlight areas have gotten darker. Also notice that Photoshop again bends the curve; the resulting curvature below the initial line causes most of the tonal ranges to be rendered with less luminosity--everything gets darker.
Now we start to work on the shadow regions. We want to add more luminosity to the shadows, so we drag the curve higher in the shadow region. The result--the shadow range is now brighter, but the curve also does the same for the rest of the tonal ranges; so the overall image is too bright now. We need to fix that.
Okay now we have something. To control the curve, we add several more vertices. Adding vertices is as simple as clicking on the line somewhere besides the already present vertice(s). Now you can see the shadow regions have been made brighter, the midtones somewhat brighter, and the highlights only just slightly brighter. The contrast now has been reduced and detail in the shadow region is now much more readily visible. This is our chosen curves setting.

Be careful when modifying curves. You can create unrealistic colors (through posterization) if you make your curves too sharp. This becomes especially true as you add more vertices to anchor the curve.

Using Adjustment Layers with Levels and Curves

One of the most power tools in Photoshop is the ability to use layers. Think of layers as being a pile of transparencies (like animation cels), each with different images. Your complete image is the total composition of these cells. You see the total composition by looking down from the top of this pile. You can add or remove layers, fix them up and put them back in. Or just simply discard.

Layers in Photoshop are as simple as that. In addition to image layers like these there are also Adjustment Layers. Instead of applying an image, they apply some sort of adjustment to the layer set. And just like regular layers, they can be removed or modified at any time without modifying the image layers. This is extremely useful; supposed you make some Levels and Curves adjustments, but later on you decide that you don't want them or you need to modify them. If you use levels you can do this; if you used the standard way you would be out of luck; you'd have to start from scratch. Even more useful, adjustment layers can be saved to disk along with your image so they can be adjusted days, months, even years later by anyone. Obviously Levels and Curves can be used as adjustment layers, otherwise we wouldn't be discussing this. Let's take a look at what they look like and how to create them.

You can see that there are two adjustment layers, a Curves adjustment layer (called "Curves 1") and a Levels adjustment layer (called "Levels 1").
To create adjustment layers, click on the button "Create new fill or adjustment layer" button. To create Levels and/or Curves adjustment layers, click on the corresponding menu item.

Adjustment layers work much like regular layers except they affect the total image; they don't have a latent image themselves. You can make them invisible temporarily by toggling the visibility button (the "eye" icon) or delete them by simply dragging them to the layer trash can. But of most interest, if you double-click o the adjustment layer (the circle that is half black and half white) you have the ability to modify that adjustment!

Using the Unsharp Mask Tool

Now onto sharpening. Inherently most images, whether taken by a regular film camera or digital one will often appear soft. Application of sharpening is often a necessary step.

Sharpening should ALWAYS be considered a final step and should never be applied to your working copy permanently. You should only use it just before making a print or just before saving a JPEG for web use. You should discard your sharpened working copy; or if you feel that strongly about it, then save it as another version of the image (maybe call it "myimage_sharpened.tif" for example).

What is Sharpening?

Sharpening is simply increasing the contrast along a defined edge. For example:
Unsharpened Image (magnified 77x)
Sharpened Image (magnified 77x)

If only implementing sharpening algorithms were that easy; they are not. There lots of ways you can sharpen images, some work better than others depending on the image. There are lots of sharpening plug-ins. Sharpening is one of those things that requires image comprehension that computers lack--it requires input from a human to fill in the blanks.

The best tool available for the job without resorting to buying third-party sharping plug-in software is the Unsharp Mask Filter. It can be found under the top menu "Filters", under the "Sharpen" submenu. Click "Unsharp Mask..." to access it.

A lot of people don't realize that and use the "Sharpen" command. That works, but you don't have any control and the results often aren't very good. Let's take a look at the Unsharp Mask tool:

  • Amount - sets the intensity or level of sharpening to apply. The higher you set it, the most contrast along edges.
  • Radius - describes the width of pixels along the edge that are affected by the increase in contrast. This describes the scale of sharpening (broad or very narrow along the edges)
  • Threshold - describes the necessary difference in tonality before sharpening is applied. This is used to define what an "edge" is to Photoshop.

 

When trying out the Unsharp Mask Filter, always zoom in to see the effect close up. This is important because if your application is too intense, your image may appear pixilated. Optimally you want to use settings that keep edges (like the definition of an eye) sharp, but areas of smooth tone (such as skin or the sky) untouched. The worst thing is to see pixilation in such smooth regions (imaging zooming into the sky and instead of seeing smooth blue across, you see lots of sharp, contrasty pixels)

Good Unsharp Settings?

That's hard to say; again it really depends on the image you are working on. You really have to play with it to get it right, and often one set of Unsharp masks settings that works great for one image may look terrible on another. I've used a variety of settings, here are some I've tried and had success with:

Amount = 200, Radius = 0.5, Threshold = 8
Amount = 120, Radius = 1, Threshold = 6
Amount = 350, Radius = 0.3, Threshold = 10

Again, you need to try yourself to come up with settings that work best.

Saving Images in TIFF Format

The TIFF format has been mentioned several times. Why use that format? Why not use JPEG or the Photoshop format?

You can, but TIFF's have some nice advantages:

  • Well known standard format
  • Lossless image preservation
  • Compression
  • Photoshop Layer Preservation
  • ICC Color profile preservation

To save a file in TIFF format, click on the "File" top menu, select "Save As..." Next you'll get the following pop up dialog:

Under "Format" choose "TIFF". Be sure to tell it to embed Layers and the Color profile (in Red) by enabling those options. Don't forget to name your file. Next click "Save" and the following pop up will show up:

Here you can specify a variety of options. To save space, you can enable the compression options, for both the actual image and the layer information. Note that the TIFF format is lossless unless you select JPEG compression--don't choose that option. A popular compression algorithm to use is the LZW algorithm. You can also use the ZIP format, however older versions of Photoshop and other photo imaging tools (not to mention the operating system) may not understand your file. Still, I choose ZIP compression and it works well for me. The other options are minor, but if you care you should probably keep the byte order according to the platform you use most often; "Save Image Pyramid" saves multi-resolution image information. For more detailed information check the Photoshop documentation or look for other resources on the web or in books.

General Guidelines

Wow that was a lot of information. Let's review, and put things in order.

  • Always keep your images color managed.
  • Never modify your originals; instead create working copies and use those.
  • Your working copies should always be saved in lossless format
  • If you need the image for the web or for making prints, make specialized versions by converting the image to the proper color space.

When using Levels and Curves

  • Use Adjustment Layers whenever possible
  • Manual Levels are more precise than Auto Leveling, but less convenient
  • Use Levels before using Curves.

When using Unsharp Mask

  • Zoom in to view the effects of your settings
  • Always apply sharpening only when preparing output versions of your image (to the web or to print)
  • Do not save your sharpened version; save it as a new version or discard it

Conclusion

Image post-processing is a vital part of photography that is often overlooked or underappreciated. A good part of your time should be spent processing your image; your images will look a lot better. There are a lot of things you can do improve your images; this article only attempts to give a small overview of what I do to fix my photographs. Color management, we have learned is a critical part of the process--ensuring that the colors we are working with stay true throughout our digital imaging workflow. It is also important we use the right file formats to prevent image lossiness. We also examined briefly the use of the Levels and Curves tool, through normal use and by using adjustment layers. We also looked at the Unsharp Mask Filter. Hopefully this will give you an idea of how to start working on your images; these simple techniques are sometimes the only techniques I use to fix up my photographs before presentation on the web or on print.

Resources

There are a great many books and websites you can find to learn more about digital imaging. Check out the following:

Books

Web Sites

Return to Front Page Back to Learning about Photography

You are visitor 295 since April 1, 2004. Last Viewed May 9, 2008 4:09PM.