Workflow Examples using a Flatbed Scanner
Before you start, you should be aware of the capabilities of your scanner, particularly what its optical resolution is. For more general information on scanners, see the scanning page.
Scanning Photos
My personal method is to scan photos directly into my photo software (Adobe Photoshop) and I'll be describing this technique. But most scanners come with software that also allows the direct scanning of an image into a file which you would later load into your photo software for editing. The actual scanning info is the same for both methods.
- I make sure that the scanner in on
- I put the first photo face down on the (clean) scanner glass
- I load Photoshop and choose File > Import. This brings up a list of items, one of which is the "twain" driver for my scanner (this would have been installed on your computer when you installed the driver software for your scanner). Of note, my current version of Photoshop (CS5) dropped direct twain support but you can get the routines from Adobe's website that support twain (and I've had no trouble with these on a 64-bit Windows 10 computer). Most other photo software programs directly support twain.
- Choosing my scanner in Adobe's import selection brings up my scanner control panel.
- In that control panel I set the scanner to scan in 24-bit color (even with a black and white photo*) usually at 600 dpi (see the scanning page for a discussion of how to choose the appropriate dpi). I set the scanning type to "photo" and I also generally set medium sharpening.
- I press the "preview" button (some scanners will do an automatic preview) and then select exactly what I want to scan.
- Once scanned, I have a choice, I can do more scans or I can exit my scanner control panel in order to return to Photoshop.
- Back in Photoshop I can see the exact results of my scan and make initial edits if required. I can also add IPTC data to the image.
- At any point, I can save my file - I usually do this first thing. I usually save it as an LZW compressed TIF file (see the filetypes page). I also follow my personal genealogy naming convention. See the How to Name Digital Photos page for ideas on how to come up with a meaningful filename for your scanned image.
That's it in a nutshell.
* I scan old photos in colour in order to capture more detail from the original image. See the greyscale discussion at the bottom of this page.
Scanning Slides
This can be done if your scanner has been designed to scan slides and negatives - it requires a light in the lid of the scanner.
- I make sure the slides I plan to scan are as clean as I can get them.
- I set the scanner up for slide scanning - removing the upper lid light cover, putting the slides in the slide holder that came with the scanner.
- I load Photoshop and choose File > Import > my scanner (see above for details).
- In my scanner control panel I choose "film" as my document type and "positive film" as the film type.
- I set the scanner for 24-bit colour and usually set the dpi to 3200 (see the scanning page for a discussion of how to choose the appropriate dpi).
- If it is an old slide with scratches and such, I may use dust and scratch reduction software (in my case Digital Ice) to try to improve the scanning results. I generally try one with and one without to see the difference.
- Then I use preview to select the first slide and scan it.
- (see the scanning page for a discussion of how to choose the appropriate dpi).
- Back in Photoshop I can see the exact results of my scan and make initial edits if required. I can also add IPTC data to the image.
- At any point, I can save my file - I usually do this first thing. I usually save it as an LZW compressed TIF file (see the filetypes page). I also follow my personal genealogy naming convention. See the How to Name Digital Photos page for ideas on how to come up with a meaningful filename for your scanned image.
If you plan on doing a lot of slides and are adept at learning new software, you might want to consider using the scanning software VueScan which will allow you to achieve the maximum quality your scanner is capable of.
Scanning Negatives
This is the same as for scanning slides exept choose "negative" rather than "positive" in the film type option.
Scanning Documents
The first question to ask is do you simply want to have an image of the document or you do wish to have editable text. If the latter, then you want to do Optical Character Recognition (OCR).
For just an image, it is the same procedure as scanning a photo with a couple of small differences. The scanning DPI is usually fine at 300 dpi and if it is a black and white document, it can be scanned in greyscale.**
OCRing Documents
There are generally two ways to OCR. Some scanners come with OCR built into their scanning software package - so you would run your scanner's software and pick the OCR setting. Some scanners come with stand-alone OCR software. In that case you run the OCR software directly.
With either method, you'll usually end up with the scanner's control panel. It is often automatically set for best results, but if not, scan using the greyscale** setting using at least 300 dpi
The main "trick" is to ensure that the text on the page is perfectly parallel to the scanner edge - this will help to ensure the best OCR accuracy.
OCRing is usually a 3-step process - 1) the scan 2) the recognition of the scanned image into text 3) the delivery of that text into the software of your choice (i.e. Microsoft Word).
There are generally 2 problems that crop up. The first is maintaining original formatting. How well that happens is based on the quality of the OCR software. When using low-end OCR software I usually just scan into pure text and reformat the document myself.
The second is the recognition of words - OCR software does make mistakes. Common mistakes would be the recognition of an i as an l, or a 1 as an l (or vice-versa). In some cases this won't generate a spelling error, for instance OCRing the word "mail" and converting it to "mall". So there tends to be a lot of close editing required with the OCRed document.
** Greyscale is 256 shades of grey (black to white). It is different than the Black and White setting on most scanners which is literally black and white (2 colours). So, even though the document may be "black and white," for full detail it should be scanned using either greyscale (256 shades) or 24-bit colour (16 million colours). Scanning in colour will capture more detail which is why I recommend it for photos. However, for documents and OCRing, greyscale, which will produce a smaller filesize, is usually just fine. See the Colour Models page for more information.
|
