Scanning grayscale photos

In this video, I'd like to show you how to capture a continuous tone grayscale image using the Manual toolset in SilverFast. And if you have been through the automatic version, you are going to see a lot of similarities but you are also going to see that we have a lot more control working in Manual mode. First thing we have done here is I have clicked on Prescan, so we could actually see the image and that's a good place to start. Now, let's talk about what some of the challenges are and what some of our goals are for capturing a continuous tone grayscale image. This is a great image to do for continuous tone scans because it has so many of the common scanning challenges and goals that we have in continuous tone grayscale images.

One of our first issues is going to be maintaining highlight detail and shadow detail. And in this particular image, the white fur and the dark fur are going to be critical portions of this image. We also of course want to have and address the issues of overall brightness and overall contrast, and with an image like this with so much detail, we want to address sharpening as well. So let's go ahead and just dive right in. First, of course we are going to click on Reflective, because this is a reflective image and we will go positive because this is all we can do here, because we're not working with transparencies.

And we can go 16 bit to 8 bit. Now, it's possible, you could just capture this as a 16 bit image, and then SilverFast will deliver a 16 bit image and then you can do editing in Photoshop. You can also capture this as 16 bit HDR which is High Dynamic Range, in which you do almost no editing in SilverFast, and you will do all the editing either in Photoshop or in SilverFast HDR program, which is an image editing program. This is a scanning course and we are going to be doing most of our adjustment during the scan.

But that's when you might want to choose one of these other selections here. So next, we will set our frame, and we are just going to do a general setting of the frame here. We want to make sure that we get that frame well inside the edge of the picture, we don't want any of the white background. And while we're doing that, let's go up to the Scan dimensions dialog and we are going to work a little bit with the Expert settings here. Notice I did this manually with my hand, so I end up a Width of 7.95 and a Height of 7.87 which is all fine. But what if we wanted it truly square? Well, we can just come in here and we can just set this at 7.9.

Then we can have the same output over here as well. So we can have 100%, coming in and out. But what if we didn't want 100%? What if we wanted this, instead, we didn't wanted it at full 7.9, we just wanted this at 6.0. That would be the largest we would ever use this image. First of all, notice that by having this little chain-link turned on here, when you make an adjustment on one, it will make an adjustment on the other, which is good. So we maintain proportion. But what this does here by setting an Output different from an Input, it means that SilverFast, using whatever Scan Resolution we set here, and we will address that in just a second, will automatically do all the math for you.

So in this case, if we set our Image Quality at 300 pixels per inch, which we will, it will deliver an image at 6x6 rather than 7.9x7.9. The other thing that you can do here is you can click on these locks, and what that does is it locks in this frame dimension here, so you don't by mistake click on it, and drag it and move it somewhere else. It's going to lock and maintain the position and the dimensions. Then let's go ahead and finish our discussion of resolution. And for a continuous tone image at high quality, 300 pixels per inch is plenty. And because we're not going to be doing any scaling up, we don't need anymore resolution because we have determined this would be the maximum output dimensions.

My goal here is to capture an image at the largest dimension and the highest resolution at which I think I had ever used the image. Later on in my workflow, I can make copies, I can down sample, or I can down size. I can save in other formats, but I can always come back to the original, very high quality image that I've created through my scanner. So, I am going to choose 300 as my maximum resolution. And then right up here, we are going to go ahead and name it. I like to use a logical name; our capture bit depth mode, grayscale, and then the resolution of the file, so I can tell a lot just by looking at the name of the image.

And then as always, I like to choose a file format that's uncompressed and high quality. It's either going to be TIFF or .PSD. So, we are all set in terms of the setting the Scan dimensions dialog box. Next, let's move forward and start adjusting the actual image and we are going to use our Densitometer tool here. You are going to see how powerful and helpful and useful it really is. First, let's go ahead, and just set some of the Expert settings here and we have really already set this in the Preferences, but we can just check. We are going to do a 8 bit, not a 16 bit in terms of the values that we see here, and 3 pixel Radius, the default in SilverFast is 2.

I like to use 3 just to get a little bit larger sample area. So we will just fine-tune that. And then, these two buttons right here, when we click on the right one, it shows us the lightest area, when we click on the left one, it shows us the darkest area. Well, by clicking on these two, this one on the right-hand side, I can actually set a sample point on that, a sample point number 1. If I click here, I can click sample point number 2 and that shows me the lightest area in the image and what K value is, and the darkest area of the image, and what the K value is.

And just to remind you, K stands for Black in terms of percentage, like CMYK. Is there any other point I'd like to monitor in this image? The answer is yes. I want to monitor the lightest area of the fur because that's going to be a critical portion of the image. Using my Densitometer, I can move this around because I floated it. I can look at where the Grayscale values are in the fur. Now, these are all in the 20s and 30s, 20s and 30s, and up here, it looks like we have got some of the lightest areas of the image. So if I hold down my Shift key, and then I click, I can place in with a Color Sampler point there.

Now if I want to remove points, which I am going to remove the shadow point for just a second here, I hold on my Shift key and then click on point number 2. There we go. So you can add and subtract points just by holding down the Shift key. I am going to go back and add that shadow point there, which becomes point number 3. And the reason why I've done that is in every image that I work on, whether it's in SilverFast or in Photoshop, I like to have my Sampler Points in numeric order based upon tonal range. My sampler point number 1 is always going to be the lightest diffuse highlight, number 2 is the next darkest, and number 3 and so forth down to the shadow.

So, this way, whenever I open up an image, I know from number 1, to number 2, to number 3 is going to go in increasing tonal value range. So, there we go! We are all set up, and let's go ahead and start making our adjustments. And we are going to use our Histogram tool first and I am going to use this histogram in concert with the K values that we see here to make our adjustment. First, just a quick review of what the Histogram shows us. It's a graph basically that shows us the distribution of grayscale on our image. I actually use histograms when I'm taking my photograph with my digital camera, so that I can see that I'm getting detail all the way from the highlight end on the right, to the shadow end on the left.

And what we are after here is really fine- tuning our highlight and shadow points, initially. So point number 1, K value of 8%. What's my goal? Well, I have got a goal of 5% white highlight. Some people like to go under 2 or 3. I've had circumstances where I have lost to 2 or 3% highlight on some printing devices. So I like to set mine at 5. By the way, if you're working in the newspaper industry, all bets are off, because on newspaper, you may have to go actually to 10%, and actually to maintain a highlight.

How about the Shadow? Notice it's at 88. Not bad, but we can do better. I am going to set my Shadow up at 95%, because I know that the printing devices that I work with can hold 95%. If yours can only hold 90, then put it at 90. But what this does is it restricts our tonal range of important data in our image from 5% to 95%. And notice that while I am making these adjustments, not only that point number 1 and point number 3 adjusts, but I can keep track of point number 2 which is the lightest area of one of the critical diffuse highlight areas that I see in this image, and around at 10% or 12% is still going to look very white in my image on the final analysis, so I am happy with that.

Other times, when I'll sacrifice the sky to get this critical area just the way I want it, you betcha! In this case, I don't have to, we will still end up with detail in the sky and in this white highlight here. And notice that I don't have either one of my highlight or my shadow coming into the data here, so I know I'm protecting all of my image data. Now, I could adjust the midtone here to adjust overall brightness and contrast, but I never do. I go to another tool and I go to the Gradation tool. Notice that what we have here is a straight line curve, that's how it always starts. And what we use this tool for is adjusting overall image brightness and then contrast.

I know that in almost any image that's continuous tone, when I actually print this image, it typically prints a little bit darker than what you see on screen particularly from midtone to shadow. So what I am going to do here, because of this area in the image here where there is lots of three-quarter tone to shadow, and up here, I want to make sure I lighten that overall just a little bit, so I'm just going to take my Brightening curve here, and I am just going to move it up may be about 4 or 5 points here, just to overall lighten the image. And you can see the impact on the screen when I do that. I can also take individual portions of the image like the three-quarter tone and lighten that specifically if I want to, but not necessary in this case.

How about for Contrast? Most people think more contrast, the better. Well, watch what happens when we raise the Contrast, even just a little bit. Do you see how the portrait of Zip gets a little bit too stark? For portrait images, typically we don't want more contrast unless it's really, really flat to begin with, in fact, a little bit less. I am going to drop this down to maybe about -4, -5. Look at the difference between plus and minus, and it really kind of just softens out the image. It's a great move to do for portrait images. For other images that have higher contrast demands like product shots or landscapes, sure! Throw a little bit of extra contrast in there.

But for portraits, back it off a little bit, and I think you will be happier. Notice it's a very subtle adjustment here along the quarter-tone and the three-quarter tone, but it's enough to really make a difference. One other thing we can do with a really interesting tool, and that's this tool, this is the selective color of grayscale. One of the things that I know about grayscale photos is that sometimes there is a little bit of color cast in there. In this case, because of the nature of the paper and the ink, there was a little bit of a red cast in this image, where sometimes when a photo fades with age, it gets a little bit of color.

I know there was some red in this one, and I also know that that red, that color cast particularly comes from age or paper, it is more of it in the three-quarter tone to shadow than there is in the highlight areas. So watch this while I am going to take this red slider, watch what happens to the image here when I move this red slider up. Do you see how it just lightens that whole three-quarter tone to shadow area in that image? Very, very nice. It's a very subtle job that you can do with the selective color of grayscale. All right! We fine-tuned our tonal range. One other thing to address here and that's sharpening.

I will tell you that I typically don't apply sharpening to the scale. When I do an image capture like this, I like to get the frame, the resolution, the tonal values all correct and then I typically like to save my original scanned photo that way without sharpening. And then I will make copies and I will selectively sharpen for particular output devices. It doesn't mean you have to do that. If you want to get SilverFast to do all the work, you want to scan the image, sharpen it, and then just go to Print, and not have to fuss with it anymore? Let's talk about doing that.

So we are going to use the Unsharp Mask tool. You will notice as you activate these tools, the active tool has that little red button on there, just a little visual clue to you. There are three key variables here, and the first one is Power. In Photoshop's Unsharp Mask tool, that's called Amount. It's set at 100%, the Radius which affects the width over which it will be applied in terms of pixels, and then a Threshold value which controls where in your image in terms of grayscale value sharpening will be applied. And it's a Threshold value of 1 is what this is set on which means that you have to have at least one difference in grayscale value between 2 pixels before the sharpening will be applied.

Well, I am going to set this up between 2 and 3 most of the time on my images, and because we've got lots of high contrast edges here, I want the subtle ones the very soft ones to not be sharpened very much. One other thing I am going to address here, particularly with this image is I am going to go to the Expert dialog box. There is all sorts of values in here, and you will probably forget what they all mean from time to time. You can always click on the QuickTime movie, or the PDF file to update your knowledge, or just to remind yourself what they mean. What I want to address here is two things; one the Sharpen up to.

By default, it's set at 100% I am going to set this down to 90, and the reason is I don't want the shadow areas to have a lot of sharpening in them because if you sharpen shadow areas, it tends to add graininess to them, and I don't want that. So I am going to set this at about 90%. The other thing I am going to do is I am going to check on my Soft shadows, that is, that it's going to prevent sharpening in cast shadow areas, such as the area right in here, the area right in here. We don't need a lot of sharpening in the shadow; we need it mainly in the highlight to three-quarter tone.

Again, you can do this here or you can do it later on in Photoshop, either one. So let's go ahead and click the Scan button, and the scanner may have to warm up a little bit. You can check the scanner status down here, and see, it's warming up. So if you click the Scan button, and nothing happens immediately, don't freak out, just look down there and you go, oh, yeah! The scanner is warming up. When the scan gets done, you just click on the Open Image button, and then SilverFast will open your image, and for my case, I open everything in Photoshop. I mean that's the way I have it set in my operating system. And notice we end up with a beautiful picture of Zip and notice beautiful nice highlights here at 13%, 14% 15%, the sky still has detail in it, right at 5% or 6% the way we set it, nice shadow details.

Look at that, right at 95%, and I'm just monitoring the Info tool here while I am doing this which is the same as a Densitometer in the scanning program. And notice how we have the nice soft portrait here, how we have got nice lightning and the thee-quarter tone to shadow. So there is our finished image and we've addressed all of the critical issues that we wanted to address in the capturing of this image.