Start learning with our library of video tutorials taught by experts. Get started
Viewers: in countries Watching now:
Review the scanning techniques graphics professionals and photographers use, while delving into workflow considerations and the advanced image-quality controls available in most scanning software. Author Taz Tally explains the core concepts, such as how resolution and interpolation affect scans; introduces the industry-standard SilverFast scanning software; and shares the settings to achieve the best results from a scan. The course also covers keeping your scanner and its parts clean and free of dust, and includes a variety of start-to-finish scanning tasks.
Earlier we discussed how images such as this RGB image that we see here a Greenwich is really a construction of three grayscale channels. What I'd like to do now is show you a tool if you can use to get a different look at all the grayscale data, and show you how to see where the data is, show you how to use that tool to evaluate the image and even to correct it and this tool is called a Histogram. We start in the upper left-hand corner with this Greenwich Village Gold image that you've seen before and we look down here at the Histogram panel, we see that there is data all the way from one end of this graph to the other.
This is a graph that shows you from highlight to shadow where the distribution of data is in your image. This is the highlight, this is the mid- tone, this is the shadow, this is the quarter tone, this is the three- quarter tone, and as you can see this is a pretty well-exposed image so we have data all the way from the highlight all the way to the shadow. Just by looking at a Histogram you can get a real good idea of how the image has been exposed and as we're going to see it can lead you in the path towards how you need to correct the image and this works both in scanning and in Photoshop.
What I'd like to do now is take a look at a variety of different images that have a different distribution of data to get you used to looking at and evaluating a Histogram. Let's look at this Cloud image and what I'd like to get you to do is be able to visually look at an image and actually imagine what the Histogram looks like. Notice that we've got white clouds but if we click on this image we see that there's data in the highlight but it maybe not pure white clouds and we look over this image we see, ooh, is there anything that's truly dark or truly black in here, truly shadow? When we look at the Histogram of the image we see, there's nothing that's exactly pure white and we've got most of our data from the quarter tone down to between the three-quarter tone and shadow and there's almost no shadow data in this image at all from about 60% to 100%.
One of the points I want to drive home here is you really shouldn't be trusting your eyes to look at and evaluate your image. The human eye does a really good job of making very detailed qualitative assessments of your image when it's side -by-side with another image if there's been a small change, but the human eye does not do a good job of making quantitative assessments of the content or the color characteristics or the tonal variations in an image. And a Histogram is a graphical, it's a visual tool that we can use to really help out in this fashion. All right, let's take a look at the Lighthouse image and notice the lighthouse image is mostly dark.
When we look at the Histogram we see lo and behold it's kind of almost the opposite of the Clouds image. Here most of the data is from the mid-tone down to the shadow. So notice how the Histogram in our evaluation of it matches the actual image itself. What do you think about this image here, the Gnarly Reflections image? When we click on the Gnarly Reflections image we see that most of the data in this one is from highlight to mid- tone, there's almost no shadow data. You look at it and the eye goes, oh that's dark, but in terms of the total tonal range it's really only a mid-tone. See the eye is fooled into the thinking that things are lighter or darker than they maybe based upon relative tonal or color values elsewhere in the image.
And finally let's look at the Cloudy Landscape image and we see that here we have a grouping of most of the data right around the mid-tone. Right from the three-quarter tone not quite up to the quarter tone, and that's why the image is by the way very low contrast. And in fact when we carefully, visually evaluate the image we see there is no real dark shadows in here and there's certainly aren't any bright white highlights matching indeed the Histogram. So just quickly going from one image to another you can see how that Histogram changes and how much valuable information a Histogram can be. Now so far we've been looking at the composite Histogram, haven't we? Let's go one step deeper.
Notice that these images are all RGB images which we know from earlier in the course means that this image has three grayscale channels in there. This Histogram that we're looking at is the composite. If we take this and we look at in All Channels View and we move this side-by-side to each image we can see what the Red, Green, and the Blue channels look like for each of the three channels. So this is the master Histogram and then we have three individual channels.
Let's just take a look at each of these images in turn. There's the Clouds image, there is the master, remember not much data from three-quarter tone to shadow and it's the same on all of them except for look at the Blue channel, do you see how the Blue channel has even less data than the other ones? Click on the top of the Histogram. Let's go to the Lighthouse image, and here we see again missing data in the master channel but all three of them have lots of missing data, but notice that the Blue channel is a little bit further to the right than the Red and the Green. And when we look at this image that tells us, then that this image has a little bit of a blue color cast to it.
Let's move to Gnarly Reflections and we see the same thing but it's even more dramatic, where we have the master Histogram, where we see data from mid-tone to highlight. When we look at the individual channels notice how the Red and the Green are almost coincidence, the Green is offset a little bit, but do you see how the Blue is offset even more indicating that this has a blue color cast to it. In fact when we look at it we see this has a little bit of blue to it or pretty significant amount of blue. And finally when we look at the Cloudy Landscape we see again not much data in the highlight to mid-tone, but the Red channel is offset a little bit more to the right than the Green and the Blue channel but not nearly as much of an offset as we saw in Gnarly Reflections in the offset of the Blue channel.
So when you get used to looking a Histograms even minor variations are small changes can add up a lot of understanding of what this image is all about and where the data is. This is significant because when we look at an image such as say Gnarly Reflections we can see it's low contrast. That's fine. We can really semi-quantitatively say, oh my gosh, look at that everything from about 60% down to 100% grayscale is missing in this image. And then we can make an assessment, do we want to keep it that way, do we like the moody aspect of the image or would we like to change it? It gives us creative control of our image.
Let's see how we might use the Histogram to help us make an adjustment. In this case of course we're working in Photoshop but the same thing we'll apply during the scan. All right, what I'm going to do is I'm going to add a Curves adjustment layer and notice that when we add a Curves adjustment layer we see the Histogram actually displayed in Curves, the same thing is in levels. But I'm showing you curves, because curves gives us much more control over the distribution of grayscale data because we can adjust the curve anywhere along the total range. I'm going to do two adjustments here just to show you how we can use the Histogram to help us adjust an image.
One, I'm going to work with the master RGB channel. I'm going to click on the shadow point here and I'm going to use the Histogram to help determine how far I'm going to move that shadow point. I'm going to move it right up to the beginning of that data. What this allows me to do is still maintain the detail in the image but increase the contrast. Notice if I go too far see how all the shadow detail gets lost. So I can visually using a Curve tool adjust the shadow point up, but not to the point where we start to lose shadow detail. So I've done a master Histogram, so we'll call this one the Master Adjustment.
Now I'm going to turn that off for a minute. I'm going to go back to the background layer, I'm going to add another curve layer, and in this case what we're going to do is I'm going to go into the individual channels. I'm going to go into the Red channel and let me just cycle through here, there is the Red, the Green, and then the Blue. Remember how we noticed that the Blue channel was offset, you put my little pointer there and go backwards. See how it's not pointing at the data anymore, the Blue channel is offset more. And in this case then if we adjust the shadow point on each of the individual channels the Blue, the Green and now the Red, it's a different point at each place on each of the individual channels.
So we'll call this the Channels Correction, and look at the difference that we get. In the master channel we did them all at once, everything just gets darker, we get increased contrast, but the color balance remains the same, there's still the blue color cast. But when we did the individual channels, notice how it would neutralize the image more, take out some of that Blue and we can go anywhere in between. My point is that the Histogram is a very valuable tool for helping us not only evaluate the image in terms of where the distribution of grayscale values are in our image, it could point us in the right direction about how to correct the image and we combine a Curve tool which allows us to make corrections anywhere in the tonal range.
As we'll see when we perform this during the scan rather than in the post-scan in Photoshop we'll actually end up with higher-quality images, because we're doing this before we actually capture the data and send it out of the scanner. So histograms and curves working together, great evaluation as well as image editing correction tools.
There are currently no FAQs about Scanning Techniques for Photography, Art, and Design.
Access exercise files from a button right under the course name.
Search within course videos and transcripts, and jump right to the results.
Remove icons showing you already watched videos if you want to start over.
Make the video wide, narrow, full-screen, or pop the player out of the page into its own window.
Click on text in the transcript to jump to that spot in the video. As the video plays, the relevant spot in the transcript will be highlighted.