Digital Coin Photography

An Update in Light of New Technology

In the two years since I first posted my look at digital photography with the Nikon Coolpix 990 digital camera, There have been changes in the technology that will affect choices of equipment but the principles of photography remain the same. The earlier page could be faulted for being too specific to the controls available on that particular camera. The intent was to point out the controls available on many cameras and suggest which ones might be desirable when selecting a camera.

This page will attempt to provide more in the way of basic information applicable to a wider range of cameras but is being driven by my acquisition of a new digital camera. The Minolta Dimage 7i was selected to correct what I considered shortcomings of the Nikon. As can be expected, there was some bad to go with the good. I miss the Best Shot Selector feature of the Nikon and the way the small Nikon lens fit in my microscope eyecup with no adapter. Also the Nikon allowed closer focusing without additional adapters but purchase of a simple close up lens corrected this loss.

Least important of the features was the image size increase to 5 Megapixels. Unless prints are made in large sizes (over 8x10), many users will not see the difference between the 3.3 and 5 Megapixel cameras. More significant is the Minolta's much larger zoom lens range providing a longer telephoto and their decision to provide macro focusing at the telephoto end of the range. Those who have read my previous page might be able to understand the effect of being able to shoot coins from a greater distance on the ability to manipulate the lights as desired. While matters of lighting will be touched upon on this page, much of the earlier discussion still applies and will not be repeated here. Other features of the Minolta that appealed to me was the ease of selecting various settings without entering extensive menus and the ease of use in manual focus and exposure mode. While my new 5 Megapixel camera is a Minolta and is recommended for those wanting a high end consumer camera, similar results will be possible with models from Nikon, Sony, Canon or other manufacturers. Those wishing to save money (5 Megapixel cameras cost about twice the price of 3 Megapixel models) will be well served by the old standard models. I do not recommend purchase of digital cameras less than 3 Megapixel size but there are many who differ from this opinion.

Basics of Coin Photography

The first point I want to make about coin photography is that you MUST use a tripod or other camera support. Yes, I know some people have some luck handholding the camera and get a percentage of sharp images but, for consistent results, some camera support is required. If for no other reason, a camera support enables you to shoot both sides of the coin at exactly the same size and angle so they can be combined into one image and look right together. Of the camera supports suitable for coin photography, the tripod itself may be the worst choice. Even when inverted to allow shooting down on the subject, tripod legs can get in the way of the lights or the hands operating the camera. A better choice is a copy stand. These are available commercially or can be adapted from left over parts of a broken photographic enlarger. Almost as good is a simple stand constructed from wood and equipped with a ball and socket head to hold the camera. This is a case where a little ingenuity will go a long way and require adapting the design to match the camera. For example, the wood stand shown in the photo below was made to fit the Nikon 990 with its off center lens module. In the photo it is shown with an old 35mm Pentax SLR which looks very off center on the base.

The dowel rod coin support uses a black felt background. The camera support is adapted from an old photographic enlarger. Light is provided by a Halogen gooseneck lamp. A wood camera support shows a second ball head at top to illustrate the possibility of using multiple holes to adjust distance from the subject. The felt was removed to show the moveable support block. Glass over an old refrigerator tray allows shadows to fall outside field of view. The dowel holds the coin above the glass to avoid surface dust being in focus. A light can be installed in the base but too strong a light will cause flare in the image.

Whichever camera support is used, we need also to support the coin being photographed and provide a suitable background for the image. The simple answer is to lay the coin on a sheet of paper and not worry about shadows or textures on the background. I prefer to rest the coin on an upright dowel rod set in a wooden base that can be slid around the larger base of the copy stand to allow easy centering of the image. In addition to throwing the background out of focus, the dowel support allows the coin to be tilted slightly to produce variations of angle to control reflections from the surface. A small piece of clay (non-sulfur!) can be used to hold the coin on the dowel or the top of the dowel can be drilled out producing a concave surface that will allow some adjustment of the coin position. The dowel technique will produce a shadow of the dowel on the black background if too much light is allowed to fall on the felt. This can be removed with post-processing of the image.

A white (or other color) background can be produced by supporting the coin on a piece of glass suspended over a white surface. I use an old refrigerator crisper tray. The angle of the light will produce a shadow outside the area behind the coin. If dust on the glass shows in the photo, a short dowel can be used to raise the coin and throw the glass out of focus. Again, any irregularity in background tone can be removed in post-processing using an image manipulation program like Photoshop or Paint Shop Pro.

Light can be provided by a Halogen gooseneck lamp available widely (Wal-Mart, Target, Ikea etc.) for about $10. One will do the job but some people like to experiment with multiple lights or use one to even out the light falling on the background. The images on this page were all made with one light. These lights match the color temperature of the camera's Tungsten or Indoor setting. Some digital cameras only provide an Automatic color balance setting which may require more adjustment during post-processing. Some photographers prefer to work outdoors in natural light. This can be quite acceptable if you have a secure place to set up coins and cameras in the back yard. Since natural light will vary with time and weather, I prefer using artificial light.

Close focus capability varies quite a bit from camera to camera but most digital cameras focus more closely than film cameras commonly owned by amateurs. Some, like the Nikon Coolpix 990 (subject of my earlier digital page) focus to a fraction of an inch without accessories. Others will require an accessory close up lens to fill the frame with a coin. The Minolta Dimage 7i will fill the frame with a sestertius size coin but needs accessories to shoot large images of denarii or smaller coins. Accessory close up lenses are rated on a scale from +1 (weakest) to +10 (strongest). On the Minolta, +4 is good for shooting antoniniani while denarii need +6. As demonstrated in our close up in this paragraph, accessory lenses sometimes introduce or worsen chromatic aberration or color fringing on bright parts of the coin image. This is generally worst near the edges of the frame so shooting the coin with a little room around may give better results than shooting tightly. It will be necessary to use trial and error with every camera to see what gives the best results.

While the full size file will make higher quality prints, this image of a Pertinax denarius needed to be sized down (left) for web use. A crop of the full size image is shown at the right. Certainly it is possible to shoot smaller coins without the accessory lenses and crop the images to eliminate the excess background. Considering the larger image produced by the 5 Megapixel camera, a cropped version of a denarius might be as large as the full size image shot with a 3.3 Megapixel model. If the intent is to produce smaller images for use on the web, there is no reason to produce huge files only to reduce them later. The difference will be seen only when making large prints. However, cameras with more pixels will allow more flexibility when shooting. An image can always be sized down for use in a low resolution mode but a small image can not be enlarged to a high quality file. A multi-megapixel camera lacking close focusing capability can produce an image which can be cropped and still retain enough detail for web use. A camera with no close focus capability and a small file size will fail in making coin images.

Digital images allow sharpening in software to improve their appearance. Most cameras apply a degree of sharpening when the image is made and many users are totally unaware that it has happened. When an image is resized down to be used on the web, It is often desirable to apply another dose of software sharpening to optimize the appearance of the image. This is illustrated above with a denarius of Clodius Albinus. Following being resized down from the 5 Megapixel original, the left image has a decidedly fuzzy look. The center image has benefited from sharpening in Paint Shop Pro 7 (using either the 'Sharpen' filter or the 'Unsharp Mask' - terms and availability will vary according to the software used). The right image illustrates what happens if we go too far and oversharpen the image. The idea with this and all other controls is to avoid the look of a 'processed' image and leave the viewer to concentrate on the beauty of the coin without being aware of what was done to make it look so good.

Another factor of importance in producing high quality coin images is depth of field. When a camera lens is focused on a specific distance, depth of field (depth of focus) refers to the areas in front of and behind that distance that are also in acceptable focus. This sharpness falls off gradually with sharpest results only at the distance focused upon. How great is the depth of field depends partially on the standards accepted as sharp by the photographer's eye and is affected by the size of print to be made, the distance to the subject, the focal length of the lens and the aperture (f/stop) used. Greatest depth of field is achieved with the smallest aperture (the hole in lens that admits light) that is available on the camera. This is represented by the f/stop with the greatest number (f/22 gives more depth of field than f/2). Depth of field is quite small when focusing close enough to fill the frame with a coin. Unless a small aperture is used, it may be difficult to get sharp focus on the high relief and flat fields of ancient coins. Photographers who shoot coins hand held usually are using wide apertures to allow the camera to set a fast enough shutter speed to compensate for their body trembles.

Our example image shows a Republican denarius photographed from an angle to show the serrate edge to an advantage. The left image shows the result of shooting at the maximum available aperture on this camera (f/3.5 at 200mm zoom). Note sharpness is only achieved over a small part of the inclined surface. Stopping down to the minimum aperture (f/9.5) extends the area of sharpness both in front of and behind the point of focus but still does not extend all the way to the back edge. Many digital cameras have very restricted aperture ranges. The Minolta will stop down only to f/9.5 at full telephoto zoom (f/8 otherwise) which is barely satisfactory for macro photography use. Care in focusing is always necessary. This problem is exaggerated in our intentionally inclined example but can be a problem with flat photos of very high relief coins or when the surface of the coin is not exactly perpendicular to the lens axis. Those who use a tripod or other camera support can stop down to a small aperture and slow shutter combination that will produce the sharpest results. The importance of releasing the shutter without jarring the camera was covered on my earlier page. The Minolta, like the Nikon, lacks a cable release socket but can be tripped with a USB accessory cable or by using the self timer. To achieve sharp photos we must have quality optics, accurate focus, sufficient depth of field, good camera support and non-jarring shutter release. If any one of these is missing, the image will be unclear.

Exposure for digital camera images is a bit like that for color slide film: It is best if correct exposure is given but it is better to err on the side of slight underexposure if correct exposure can not be given. The opposite is true when using color negative films which respond better to error in the direction of overexposure. With digital images, much can be done while using post-processing software but highest quality will require the original exposure to be at least close to correct. In addition to exposure, digital image files allow working with image contrast and color saturation settings to produce the best looking image. This is a matter of learning to use the software and training the eye to recognize a better result as opposed to a merely good result.

To illustrate post processing the image file from the camera requires two images. On the left is a dupondius of Nero which shows 'normal' contrast and density only in the center of the image. The left of the image is lighter than natural while the right is darker. Similarly, normal contrast is shown in the middle of the image while the top is higher contrast and the bottom is lower contrast. The image, then is actually divided into nine areas ranging from light and flat (lower left) to dark and contrasty (upper right). My definition of 'correct' may differ from yours. The right image of an antoninianus of Probus shows a similar division based on color balance. Normal is in the center surrounded by color variations in the direction of three primary and three secondary colors. Recognizing fine differences in color balance, density and contrast requires a bit of practice. The variations in the examples presented here are rather coarse compared to what you will want to accomplish when adjusting your actual images. To repeat myself: Opinion will play some part in deciding exactly when to stop manipulating the image and accept the file as the best possible.

It is important to note that many digital cameras have a small screen on the back that shows a "What You See Is What You Get" image that will allow us to evaluate the lighting of the image before we press the shutter. The image in this paragraph illustrates the great difference we can get simply by moving the coin or the light a small amount and changing the glare or reflections as seen from the camera position. Which is better may be a matter of opinion. The left version shows the rough surface of this little coin rather accurately but the right version blanks out the fields and emphasizes the legends. The point to be made here is that each photographer must learn to look through the camera and see what can be done to produce a pleasing image. While I could give a few general hints, I prefer to point out that every coin needs to be looked at as an individual and deserves to be wiggled around in the light until the result is the best we can achieve. Some coins will be easy to make look good while others will require a dozen reshoots before we get a good result. Many coins will shoot well with the light at 45 degrees from the top and fields exactly parallel to the film plane. Some will not. Learn by doing ... and doing again ...and again.

Scanners vs. Digital Cameras, a Postscript

Previously, I have posted a page on using a flatbed scanner to produce coin images. Dollar for dollar, there is no doubt that the low end consumer grade flatbed scanner is the best buy for coin imaging. Beautiful images can be produced of 90% of ancient coins for a small fraction of the price of a digital camera. I remain of the opinion that a scanner is the first accessory that should be purchased by every computer owner. The worst problem with the use of a scanner for coin images is exactly the same as its greatest strength. The light of the scanner moves down the bed maintaining a constant angle with the pickup device. This means that the angle of lighting is set by the device and is not adjustable by the user. The good news is that the novice user is saved from having to make decisions on lighting. The bad news is that the scanner can not be adjusted to improve the lighting angle when the coin being imaged is not one of the 90% suitable for that standard angle. This statement is shown to be not quite true by the image above. The same coin imaged with the digital camera in my last paragraph above was scanned three times making a slight change in the direction of the coin on the scanner surface. The center image was scanned as you see it here but the outside coins were rotated in software to restore 'up' to the proper position. This minor rotation produces quite a difference in the results with this difficult subject coin. Black bronzes with less than perfect surfaces and less than subtle cleaning are the hardest coins to image. I am not presenting this as an example of perfection in scanner technique but merely offering it for comparison to the camera image of the same coin. Which is better? Both have their strengths and weaknesses. The moving light of the scanner produces an even illumination impossible to duplicate when using a camera. The scanned image will rarely be troubled with glare. Quality scanners with CCD pickups will not be troubled by depth of field problems or camera shake. Compare the images offered here and visit my flatbed scanner page for some other examples. My opinion is that the coin imager needs both tools to be prepared to produce good images of every coin.

I will end this page with the image that I first posted in 1997 just a few days after I first used a scanner for coin imaging. I still believe it is a beautiful image and captures the beauty of the subject coin quite accurately. Could I make as good an image of this coin with my 5 Megapixel camera? Possibly. That possibility will not drive me to stop using the scanner for coin images when the subject would benefit from its characteristics.

Return to main page