Thursday, January 21, 2010

Getting the most from your digital camera

(SOME OF STUFF ABOUT CAMERA COSTS AND CURRENT FEATURES IS DATED, I WILL UPDATE IT SOON)

There are a number of things that must be set prior to taking a picture with your digital camera. These settings will make all the difference between very nice digital images that require little or no work before printing and very poor images that no amount of digital image manipulation can restore to a usable state. Most digital cameras have many more settings than even the most advanced 35mm film camera. While the initial setup of your digital camera may seem tedious, the majority of the settings will seldom need to be changed. 

Resolution


The first thing that should be set is the number of pixels that your camera will record. This is often referred to as the resolution of your camera. Even though your camera may be able to capture 4 million pixels, it doesn’t mean that you have to capture that many. Most digital cameras allow you change the number of pixels that the camera will record. If you have a camera capable of capturing 4 million pixels, it is also likely to have setting that allow you to capture only 2 million, 1 million, or .3 million pixels. This allows those who do not always need 4 million pixels of information to capture closer to the amount of pixels they need. If you only need to make a 4x6 images and you are not likely to crop them, then capturing only 2 million pixels makes more sense, as each image will be substantially smaller in file size. This will allow you to fit more images on a card, take images more quickly, and transfer and work on images more quickly. If the intended application is for screen use only, uses in which the images will appear on screen only such as PowerPoint, e-mail, or on a website, then .3 million pixels is more than adequate. Using this setting would allow you to take a number of images and then e-mail them directly from the media card without having to open them all up in an image-editing program to make them smaller.
If you are unsure of the future application of your images, then be sure to capture the greatest number of pixels your camera can record. After all, the price you paid for your camera was largely tied to the resolution of your camera. You might as well get the most from your money.  It is relatively easy to discard detail, but is impossible to add detail. So, I suggest that you set your camera to record the greatest number of pixels it can capture unless you have a good reason to do otherwise.

White Balance


The vast majority of film is of a type called ‘daylight balanced’. It is designed to be shot under lighting that is daylight balanced. Daylight balanced means light that is near the color of daylight at noon. The color balance of light is measured in degrees Kelvin. Daylight film is balanced for 5500K. Artificial lighting such as tungsten bulbs and fluorescent tubes put out light of a much different color than daylight at noon. Tungsten bulbs put out a light balance close to 3200K. Tungsten bulbs emit a very orange light.  and most fluorescents emit a very green light. The reason that everything does not look off color to us when we are indoor light is that our brain can quickly adjust to varying light conditions so that things look normal. Film cannot.
You have likely received prints back from you photofinisher that appear very orange or green. This is a failure on the part of your photofinisher. It is their task to correct for colorcasts introduced by artificial lighting. You can also use color filters to correct for the most common types of artificial lighting. When using transparency films, you have to make use of such filters, as there is no successive photofinshing steps in which color correction can be performed. Refer to figure 1 for a list of filters to use when shooting under artificial lighting.

80A     Converts daylight film for use with 3200 K lamps                blocks 2 stops of light
            For use under Tungsten bulbs (household screw in bulbs)

80B     Converts daylight film for use with 3400 K lamps                blocks 2 stops of light
            For use under Quartz bulbs (Studio lights such as Smith Victor lights)

FLD    Converts daylight film for use under fluorescent lights         blocks 1 stop of light

The greatest drawback of color correction filters is that they block a lot of light. Often the indoor light is low and the use of a color correction filter may necessitate an excessively slow shutter speed or a larger aperture than desired. When shooting print film, you can choose not to use a filter and rely on your photofinisher to color correct when printing. With a digital workflow, you would then have to color correct the image after scanning. Color correction is both laborious and time consuming. It is always better to make the extra effort to color correct via filtration rather than to color correct the digital image. However, that is not always possible. When shooting slides, color correction must be achieved through the use of filtration, as there is no subsequent color correction unless the slide is scanned.

Unlike film, digital cameras allow you to match the color reproduction to the color temperature of the light source. Degree of control is dependant on the particular model, but most have a number of presets for common light sources such as Tungsten, Fluorescent, Daylight, Flash, and Cloudy. Some even allow you to further tweak each of those settings to compensate for slight variances in the particular light source you are photographing under. This allows you to quickly set the color reproduction to the light source.
A really nice feature that is becoming more common is the ability to set the white balance manually by taking a reading off a gray card (or white card with about -1.5 stops of light dialed in) that serves as a white balance target. Using this method, white balance can be achieved in very mixed or unusual lighting situations. This feature is often called “custom white balance”. If you do not have such a feature, and must shoot under mixed lighting conditions such as daylight streaming into a room lit with fluorescent lighting, then color correct for the dominant light source.
Insuring that white balance is accurate is extremely important. If the white balance is not accurate, a colorcast will be introduced. This colorcast must be removed in an image editing application such as Photoshop or Photoshop Elements. Correcting color can be very time consuming. Insuring that you have the white balance set correctly insures that you have to spend the minimal amount of time in front of the computer.
Most digital cameras offer an auto white balance mode that works well in most circumstances. Though you should turn it off if you are trying to get a certain effect due to the color temperature of the light. For example, if you shoot pictures of fireworks, then you should set the white balance to a Daylight setting. Using the Auto mode, the camera will try to make the color of fireworks white rather than recording the colors of the fireworks. Also, many photographers like to take pictures during sunrise and sunset due to the warm light quality of the light. If you set white balance to auto, the camera may to try to make the light appear neutral in color rather than the desired warm light. In this case you should set your camera to the Daylight setting.

IMPORTANT NOTES
As digital photography requires the use of a computer in order to get the most of our images, it is assumed that you have at least some limited experience using a computer. As most people utilize the Windows operating system, references are often made to it. 

Key concepts:


Pixel-an abbreviation for picture element; it is the smallest building block of a digital image. The pixels in a photo are laid out in a grid like tiles in a mosaic. If you multiple the number of rows and columns of sensors given as a figure such as 640x480, it gives you the pixel count of 300,200. If you then multiply that number by 3 you get the largest uncompressed file size that the camera can produce. In this case a camera with an image sensor of 640x480 could produce a file of around 1 MB file (921,600 bits). You can use this to evaluate your memory needs. 

Megapixel- a marketer’s way of saying One million pixels. Any digital camera that you buy from here on out should at least be 1 megapixel. Cameras with over 3 megapixels are now available for less than 700 dollars. At the moment, 5 million pixels is the highest pixel count available in consumer digital cameras for less than $2,000.

Pixel count-Often referred to as the ‘resolution’ of the camera, the maximum pixel count is an indication of the level of detail that a camera can record. Pixel is an abbreviation for picture element; it is the smallest building block of a digital image. The pixels in a photo are laid out in a grid like tiles in a mosaic.
The more pixels that make up an image, the more detail captured. Imagine that you have been paid to make a mosaic of my dog. My dog is a black and white (harlequin) Great Dane. First you are asked to do so using only one tile. With only one tile, you can’t indicate that my dog is both black and white, so you select a black tile since my dog is predominately black. Any viewer of your mosaic would only be able to conclude that my dog is black (or at least predominately so). If I were to give you two tiles to make the mosaic, you would likely select one black tile and one white tile. Now the viewer of your mosaic can tell that half of it is predominately black and half of it is predominately white. Already, you discover that the more tiles you have to work with, the more detail that can be shown. Now use seven tiles to make a mosaic of the dog. Alternate between white and black tiles. Now the viewer can tell you more about the dog in terms of its subtleties in coloration. Lastly, divide each of the tiles of the previous drawing into quarters. Now you have a mosaic of the dog drawn with 28 tiles. Fill some of the pixels in with black that were white and erase the black from some that were filled in. Now the viewer can ascertain much more information about the dog than they could of with only 1, 2, or 7 tiles. This is what pixels are all about. The great the number of pixels used to form an image, the great the detail that is recorded. That being said, there a numerous other factors that affect overall image quality.

Resolution- Image resolution refers to the number of pixels (tiny blocks of color or gray shades that make up a picture) available to represent detail in a given area. A photo with more pixels per inch is said to have a higher resolution than a photo with fewer pixels per inch. Higher resolutions allow for more detail and subtle color transitions in an image, but they also mean bigger file sizes (not a huge drawback). Resolution can be thought of as an indicator of the scanners ability to record detail (There are many other factors that affect how much detail a scanner records such as level of compression applied, quality of the lens, focus, and accuracy of exposure.  Think of it this way, the more pixels available to render a subject the more detail that may be available. If I were to draw a picture of my dog using only one dot, you wouldn’t be able to tell much about my dog. If I were to draw a picture of my dog using 100 dots, then you would be able to tell me much more about my dog. 

Resolution for scanners is often stated something like ‘1600x3200 dpi resolution’. It is the first figure of 1200 that really matters. It is the optical resolution of the scanner. It means that the scanner is capable of making 1200 distinct samples per inch along the width of the scanner bed because there are 1200 sensors per inch. The second figure of 3200 is number of samples per inch that the scanner is capable of making along the length of the scanner by making very tiny movement of the scanner along the length of the bed. The problem is since the scanner can only make 1200 samples per along the horizontal axis, it has to use interpolation to make up information so that you have 3200 by 3200 samples otherwise you would end up with rectangular pixels which is not an option.

ONCE YOU GET YOUR IMAGE INTO ADOBE PHOTOSHOP

First, from the main menu go to Image > Resize > Image Size. We want to know how much information we have. There is no sense working on an image that has only 640x480 pixels if what you need is a stunning 16x20. If you have an image of 8x10 at 300ppi, but your only intent is to have a small 2x3 at 72ppi for use on Ebay (an online auction site), there is no sense in working on it at its original size as that will only lead to your computer running sluggish. In short, it is best to know what you have before you begin working on it. After you are done with that, look at all parts of the image at high magnification to check the quality of the work. This way, if you see a dust or hair on your image on a very detailed part of the image, you can simply rescan the image without the dust (Assuming that is the source of the image). You also want to make note of elements you would like to change. Perhaps you notice a telephone pole growing out of someone’s head or someone has a pimple you would like to remove. Once you have assessed the situation, adjust the tonal values using Levels. Then adjust the color using one of the many color tools such as Color Cast, Variations, or Levels.  Take care of the issues that you decided are necessary. Then sharpen your image.  Do you have a lot of fine dust or scratches? Then consider using the Dust and Scratches filter. Once you have everything else taken care of, Save the image using the Save As under Edit making sure that you name it something relevant. Only then do you crop the image as desired. You can choose to save this cropped version, but it isn’t necessary as you can always re-crop it later from the final un-cropped version. Then print it.

Interpolation- a method of generating a larger file size by using various algorithms to infer information between existing pixels in order to generate additional pixels. This allows one to make a larger print without the appearance of jagged looking lines caused by caused by pixels becoming so large that their square shape becomes visible.  Interpolation generates more information, but it cannot add more detail. It makes a best guess based on what information is available. Great for when you really need it, but only for when you really need it. Scanners (and some digital camera manufacturers i.e. Kodak, Sony, and Fuji) will often list an interpolated resolution such as 7200x7200. Do not base your purchasing decision on this. Look only at the true resolution, often referred to as ‘effective pixels’, given such as 600x1200. The quality of the interpolation depends largely on the method and quality of the software you are using.  Thus, you should use your image editing software (Photoshop Elements) to apply interpolation rather than your camera or scanner because it is more likely to do a better job. Remember, you only want to do it when you really need to. How will you know if you need to do it? Make the desired size print, let’s say an 11x14, and then view it from the distance it is most likely to be viewed from. Are the ‘jaggies’ visible? If they are, then interpolate the file to a higher resolution. After the interpolation, the image will look softer than it had before. You will need to sharpen the image before printing because interpolation invariably causes the image to appear soft. Then make another 11x14. Are the ‘jaggies’ less visible or gone altogether? Assuming they are, are there an artifacts left from the interpolation process? Areas of solid color sometimes appear mottled. Which of the two images do you prefer? I have found that the additional process and huge file that are created as a result (a 2MB file may balloon up to 20MB) don’t generally justify the effort. That being said, a well-exposed digital camera file will interpolate better than a scan from film or print because of the lack of grain. 

Compression-a method to decrease a file size by means of applying an algorithm to code the information.  Most methods of compression cause degradation of the image due to inaccuracies in the uncompressing (decoding) of the information. Saving an image as a JPEG file compresses the image. An image that was compressed and uncompressed will not have a lower pixel count, rather the accuracy of the reconstituted pixel compared to the uncompressed pixels will suffer. A little bit of compression is great because it allows us to fit a great deal of pictures on a relatively small capacity media device with little or no loss in discernable quality. Too much compression leads to obvious pixelated (jagged lines) images with poor tonal gradations and little detail. This type of compression is know as ‘lossy’ as image  information and thus detail are lost. There are other types of  compression though called ‘lossless’ which does not cause entail a loss of detail or quality. However; it is not the most effective. The dialog box for saving an image as a TIFF will ask you if you want to use LZW compression. This is a form of lossless compression. Let’s say you have a file that would be 20MB as a TIFF, if you were to allow the LZW compression to be applied it might be around 17MB. I know that does not seem to be a lot, but I have used it a lot to squeeze one more image onto a ZIP disk. The only drawback to using it is that additional time is required to open and save TIFF images using LZW compression.
Unless you have limited storage or you know that you only intend to use the image for screen use, than do not compress your images.

Shutter lag-The delay in time between when you press the shutter release button and when the camera actually takes the photo. All but the very most expensive digital cameras suffer from this problem. Film cameras also experience shutter lag, but it is generally so short a time as to not be noticeable. The greater shutter lag that digital cameras experience is partly due to the fact that the sensor has to be ‘zeroed’ out immediately before each shot.  This takes some time. In addition, the camera has to perform white balance operations that film cameras do not perform. Lastly, the camera has to autofocus and determine the exposure prior to taking the photo. This is a lot of computational processes taxing a small processor. One thing that you can do to insure that you capture the moment you want  is to lock the exposure and autofocus in by holding  the shutter  release halfway down prior to the  moment  you want to take the photo. Thus the camera has less tasks to perform when you press the shutter the rest of the way down to actually take the photo. You could also try to focus manually instead of relying of the camera’s autofocus. You can also use manual exposure mode and set the exposure prior to taking the photo. Lastly, you can manually set the white balance prior to taking the photo. Each of these will minimize the tasks the camera has to perform prior to taking the photo.
Cameras today have much less shutter lag than they did even two years ago. If you try the tips above and still find the shutter lag intolerable, then check some newer cameras and see if they would meet your needs.

Burst rate-The number of frames that your camera can take over a given period of time (generally measured in frames per second). My film SLR can take 7.4 frames per second. My digital SLR can take 3 frames per second. 

Burst Depth-The number of frames that your camera can maintain the burst rate for. Digital cameras have something called a ‘buffer’ that is akin to RAM in a computer. When you take a photo, the image data is temporarily held in the buffer while the process of writing the data to the card begins. When you take another photo in rapid succession, this image is also held in the buffer. Eventually, the buffer becomes full as information placed there at a rate faster than the rate it can be written to the card. This effectively limits the number of frames you can take in rapid succession. The burst rate for my digital SLR is 3 frames per second, but the burst depth is 6 frames. This means that I can take 6 photos at 3 frames per second. After that, the buffer is full and I must wait a number of seconds for the buffer to write enough data to the card so that there is room enough in the buffer to hold another image. Film cameras do not have any such restraint on the number of shots that can be taken other than the number of shots of the roll of film. You could say that my film SLR has a burst depth of 36 frames as that is the number I can take at 7.4 frames per second until I have to rewind the film and put another roll in. This issue of burst depth is a major issue to those such as sports photographers and wildlife photographers who need to be able to take a series of photos without interruption. If you find that you need a greater burst depth, then you can lower the number of pixels that the camera will record. Since the lower number of pixels recorded will result in small image files, more images can be held in the buffer resulting in a greater burst depth.

White balance: Unlike film, digital cameras allow one to match the color reproduction to the color temperature of the light source. Degree of control is dependant on the particular model, but most have a number of presets for common light sources such as Tungsten, Fluorescent, Daylight, Flash, and Cloudy. Some even allow you to further tweak each of those settings to compensate for slight variances in the particular light source you are photographing under.   Perhaps of greatest benefit is the ability to set the white balance manually be taking a reading off a gray card (or white card with about +1.5 stops of light dialed in) which serves as a white balance target. Thus, white balance can be achieved in very mixed or unusual lighting situations. If the white balance is not accurate, a global color cast will be introduced that will be maintained right through to the final output of a color managed system unless successive steps are taken to correct for it in an image editing application such as Photoshop or Elements. Most digital cameras offer an auto white balance mode that works well in most circumstances. Though you should turn it off if you are trying to get a certain effect due to the color temperature of the light. For example, if you shoot pictures of fireworks, then you should turn it off as the camera will try to make the color of fireworks white rather than recording it in the colors that you see and want to record. Also, many photographers like to take pictures during sunrise and sunset due to the warm light that the sun emits at those times. If you set white balance to auto, the camera is going to try to make the light appear neutral in color rather than the desired warm light. In both cases, you would be best served by setting your camera to the Daylight setting.

Exposure: The number one thing you can do to improve your pictures is to get the exposure right. This is particularly true with digital cameras as they only have a very narrow range of tonal values that they can record detail in. This is to say that digital cameras can only record detail in a very narrow range of brightness values. If you have ever shot slide film, you are probably aware of how hard it is to capture detail in the very bright parts of a scene while also capturing detail in the very dark parts of scene. Negative (print) film is the best at recording detail over a broad range of tonal values. Digital is the worst. Thus, it is very important that you get the exposure right. If you don’t get the exposure correct with a negative film, you have some degree of leeway as the subject tones were still probably be recorded somewhere on the film, but you will end up with less shadow or highlight detail. With digital, if you don’t get the exposure right, you will end up with little or no shadow or highlight detail. When film is grossly overexposed or underexposed it leads to a color cast due the differing color reciprocities of the 3 dye layers which make up the film emulsion. Gross over or under exposure can also result in images taken with digital cameras having a color shift. Harder to correct is noise generated in areas that are underexposured. In addition, when you shoot negative film, it is your photofinishers’ job to correct for your exposure errors as best he/she can. With digital photography, you are your own photofinisher so must take the time to improve any images with incorrect exposure.
When shooting with a digital camera, you base your exposure on the highlights. This is the opposite of shooting with print film in which you base your exposure on the shadows. Print film is very tolerant of overexposure. Digital cameras are very tolerant of underexposure with little or not tolerance for overexposure. With digital, overexposed highlight areas can not be saved. When shooting digital, you set your exposure to allow in as much light as possible without blowing out the highlights. Visually inspect the image using the playback and playback zoom feature, analyze the histogram, and/or check the blinking overexposure/hightlight indicator to determine the accuracy of your exposure. We will be spending considerable time learning to make maximum use of your camera exposure tools. The histogram of an overexposed image will show a pile up of data along the right wall. If this is the case, then the image is too bright; i.e. overexposed. Using the exposure compensation feature, you would try something like
-1.0. If still too dark, try -2.0. These whole stop changes will often overshoot the ideal exposure so you may want to try fractional stop increments such as -0.3 or -0.5. If the image is too dark, you need more light. Use the exposure compensation feature to allow in more light. You could also use your camera in manual exposure mode to allow more or less light in as needed.

Histogram:  An exposure review tool that allows one to quickly determine if an image is correctly exposed. It is a type of bar graph. The axis from left to right represents the tonal values from pure black to pure white. The axis from bottom to top represents the number of pixels. So, the histogram shows you the distribution of tonal values in the image. When there is a spike on the left wall, it indicates that some information is being recorded as pure black. Since pure black can not hold any detail, you know that some parts of the image are underexposed. When there is a spike on the right wall, it indicates that some information is being recorded as pure white. Since pure white can not hold any detail, it indicates that some information is being recorded as pure white. When shooting digital, you want to avoid any spikes on the left or right wall.

Blinking Overexposure Indicator: A fantastic exposure review tool in which the areas of overexposure blink. Since the dynamic range of digital camera is limited, you will not always be able to capture detail across the full tonal range of the scene. Knowing exactly what areas are overexposed allows you to determine whether or not to allow those areas to remain overexposed in order to capture more shadow detail.

Color mode/color space: Higher end digital cameras allow you to select a color space for the images. Usually there is a choice between one or more sRGB color spaces and Adobe RGB. Adobe RBG has a larger color gamut than sRGB colorspaces. It is also the colorspace that I suggest most people should use a working space. Thus, if you captured the images in Adobe RGB colorspace and then brought them into Photoshop you could work on them in Adobe RGB colorspace. This way, you do not have to do any colorspace conversions which invariably result in lost data. Thus, the only colorspace one would ever have to work in would be Adobe RGB. The printer driver would handle the conversion to CMYK (This paper is dealing with printer outputs designed for hard copy final output (i.e. prints for sale/display) such as desktop inkjet printers which essentially function as an RGB device, and printers such as the Fuji Pictrographs which are truly RBG devices so the issue of CMYK conversion is unnecessary).
One of the issues to be addressed is the embedding of these colorspace profiles into the image file. Some cameras support this, others don’t. Thus, if you shoot something in Adobe RGB colorspace but the profile isn’t embedded, then you have to assign the profile upon opening the file in Photoshop.

Minimum close focusing distance- The minimum distance from a subject at which the camera can still focus. This is very important to anyone with a penchant for close up or macro work such a pictures of a baby, flowers, insects, jewelry etc… In general digital cameras really have traditional film cameras beat in this department. It is not uncommon for digital cameras to be able to focus up to 2 inches away from the subject. It is often necessary to put the camera in a special macro mode to get the best close focusing ability of the camera. This mode is often illustrated with a small flower icon that looks like a tulip in silhouette.

Digital zoom-often misrepresented as something it isn’t, ‘digital zoom’ should really be called ‘in camera cropping’. The center area of your image is simply shown at a high magnification. The outside area is cropped off. If you use a 2X digital zoom, you end up with an image with only half the resolution it would of captured had you not used the zoom. That is because half that information was thrown out when it cropped. It simply crops down to a narrower view without adding any additional image information.

Focal length-technically speaking, this is the distance in mm from the nodal point of the lens to the film plane. Practically speaking, it is indicative of the lenses field of view. It is dependant on the size of the sensor and design of the lens. As the size of the sensor in most digital cameras are very small compared to the image size of 35mm film, the focal length of digital camera lenses are usually stated in terms equivalent to the focal length of a 35mm film cameras.  This is because many people have a sense of 35mm lens focal lengths.
The field of view covered by a camera using a 50mm equivalency will be just about the same as the field of view that we see with our unaided eyes. Thus it yields a picture with perspective very close to how it actually appeared. As the focal length decreases, the angle of view becomes greater. A picture taken with a 25mm equivalent will have twice as much of a given scene in it than that taken with a 50mm lens; however, what appears in the scene will be half the size of the what it was is in the scene with 50mm lens.  Likewise, a picture taken with a 100mm equivalent lens with yield a picture with just half of given scene as compared to the picture taken with the 50mm lens, however what is in the scene will appear twice as large. So, a lens with a 35mm equivalency or less will give a wide angle of coverage so it is referred to as a wide angle lens. Great for landscapes and large scenic view (the Grand Canyon, New York City skyline, etc.), photography within tight confines such as in the house, and great for group shots of the family, but terrible for single person portraiture and pictures of items which need to appear accurate in scale as wide angle lenses have a natural tendency to distort the apparent space and size relationships of items.
Lenses of 50mm are often referred to as 'normal lenses' as they render things fairly normal to how they appear with the unaided eye. Lenses of focal length of greater than 50mm are referred to as telephoto lenses or often referred to as 'long lenses' as there length often increases as the focal length does. A lens with a focal length between 70-135 are great for portraiture of a single individual or small group. It allows some working distance so that you do not necessarily have to invade the subjects personal space while still yielding a picture which looks very natural.
Lenses over 200mm are often referred to as supertelephotos and would be used by nature photographers, sport photographers, or anyone else who wishes to capture an image from afar. Typically, a digital camera that doesn't have a zoom lens will have a wide angle lens of around 35mm. Great for pictures of large areas, houses, landscapes, but not for pictures of just a few people or less. Next, you will find a group of digital cameras with a small zoom of approx. 35-70mm. This is a very handy range and will meet the needs of most. From here, you get zooms that typically range from 35-135mm. An even more handy range which incorporates an even greater telephoto ability which will be appreciated by those who like a little more working distance from their subject. Focal lengths can often by increased or decreased through the use of auxiliary lenses with some degree of image degradation or distortion.

Image stabilization- a new feature for digital cameras, but a standard in video cameras. A technique to reduce vibrations that can often ruin a picture. Two kinds exist: optical and electronic. Optical is much better as it doesn’t reduce the quality of the image nearly as much as any electronic means. Most benefit is gained when using long focal lengths that generally call for long exposure times, thus you will see this feature only in cameras with long focal lengths (at least for the moment).

Dynamic range- a scanners/digital cameras ability to capture detail across a range of brightness values. Scanners’ dynamic ranges are usually provided as a number. The minimum and maximum tonal values are called the Dmin and Dmax respectively. A film’s entire tonal range is measured by subtracting the min from the max, like a Dmax of 3.3 minus a Dmax of 0.3 means an overall density of 3.0D. Negative film is rated around 3.0D, while prints are usually less than 2.0D, and transparencies or slides are as high as 4.0D. With film having a theoretical maximum of 4.0, high-end film scanners will generally range from 3.6 thru 4.0. A new Nikon scanner has dynamic range of 4.2! Entry-level flatbed scanners are not likely to tell you their dynamic range or Dmax. Higher end flat bed scanners’ dynamic range generally ranges from 3.0 thru 3.4. Within a given price point, look to see which scanner has the highest dynamic range/Dmax. Most manufacturers use the term Dmax and dynamic range interchangeably even though this is not technically correct.
The dynamic range of digital cameras is very poor when compared to that of film. Perhaps this is why it is not usually stated. Generally speaking digital cameras can only capture information within a range of 5-6 stops.

Bit depth- (also referred to as color bit depth or color depth) is a measurement of how much information can be assigned to each pixel. If you have a very low bit depth, then you have limited means to describe a pixel. The greater the bit depth, the greater the range of values the device has in assigning to the pixel. In real world terms, it gives the device a greater ability to differentiate between subtle differences in colors and brightness values. A 24 bit depth is the norm for most digital cameras. That is 8 bits per channel for each of the three primary color channels; red, green, and blue. Unless the camera has a RAW file format which captures information in a greater color bit depth, you can assume the bit depth of a consumer oriented digital camera is 24 bit depth.  The increasingly common RAW file format allows a for the creation of a file with greater bit depth than 24. Generally, they support 14 bits per channel, making a total bit depth of 42. As Photoshop doesn’t offer a 14 bits per channel workspace, RAW file format images will be saved into a 16 bit per channel workspace even though there is only 14 bits per channel of information. 
The greater bit depth that RAW file formats support really makes a big difference in terms of overall image quality. Color and tonal gradations are much smoother in high bit depth images. There is a real trend towards high bit depth images. Unfortunately, the support of high bit depth images is one of the few features that Photoshop v.7 has that Photoshop Elements v.2.0 lacks. That does not mean that you cannot take advantage of high bit depth images though. Most digital cameras that allow you to capture high bit depth files come with a program or Photoshop Elements Plug-In that allows you make the conversion from high bit depth images to 24 bit color depth which Photoshop Elements supports.
GIMP doesn’t yet support 16 bit per channel images. However, if you are shooting RAW files, the editing that occurs to the RAW file occurs in a 16bit per channel working space which means most of the benefit of high bit depth is realized.

Noise-A random pattern of inaccurate pixels that are created when the scanner or digital camera fails to record detail in darker areas of the image. Most often seen in shadows or dark areas, it is also found in underexposed pictures. All sensors are limited to what they can sense, when that is exceeded noise is generated. Noise is akin to static on the radio. If you set your camera at a higher ISO (film speed) equivalent than the normal mode, you are more likely to have noise in the image.

Exposure compensation- a feature found on almost all digital cameras which allows you to take in more or less light than the camera would on its own. For example, you may want to take a picture of a person that is strongly backlit. The camera doesn’t know what you want to take a picture of, so if it determines that there is X amount of light present, then it will make an exposure based on that. The resulting picture will likely have a properly exposed background with a silhouette of the person. One option would be to force the flash to fire to illuminate the person while the camera makes an appropriate exposure for the background. However, this is not always an option as the person may be out of the flashes range. Using exposure compensation, you can make the camera take in more light than it thinks it needs.  You can make it take in X and 1/3 the amount of light to get an appropriate exposure of the person.  It may take some experimentation to get the amount of compensation correct, but since you are using a digital camera, this is not much of an issue.

CALCULATING FILE SIZE

NECESSARY MATH
-When calculating the size of an image to appear on screen (to put in a website, send as an e-mail) use 72ppi as this is the most common resolution of computer monitors
-In order to get a rough idea of the largest file size your digital camera is capable of generating, multiply the pixel count by 3. Thus if you have a 3.34 megapixel camera, it can generate a roughly 10MB file.

UNNECESSARY MATH
A bit is the smallest unit of data in a digital system.
A byte is a series of eight adjacent bits treated as a singe unit.
A kilobyte is 1024 bytes.
A megabyte is 1024 kilobytes.
A gigabyte is 1024 megabytes.

Most digital cameras record 24 bit color. Each pixel in a color image is made up of 8 bits of information per color channel. One each for red, green, and blue. These bits are what describe the hue, saturation, and brightness of each color. Since 8 bits equals a byte, we can say that for each pixel there is 3 bytes of information. If we take a picture with a resolution of 640 by 480 we have 307,200 pixels making up the image. Since we have 3 bytes per pixel, the total image size is 921,600 bytes. If you divide this by 1024 to get the number of kilobytes you get 900. If you divide it again by 1024 to get the number of megabytes you get .879 megabytes.

 

CONNECTIVITY OPTIONS

Getting your images from your camera to your computer can be a very time consuming process.  Cameras to computer connections are listed in order of speed beginning with the slowest. If you find that you are spending too much time with your camera downloading images to your computer, you should invest in a card reader. They offer the fastest transfer rates.

Parallel port-if you have an older printer it likely using your parallel port, as may some old peripheral devices such as a Zip drive. Old technology. No digital camera that I am aware of uses this connection. Some older scanners do. Very problematic connection. Computer must be rebooted if device wasn’t on when computer was turned on, often necessary to connect two or more devices in a series that causes recognition problems. Really shouldn’t need to buy anything using this connection anymore. Most newer devices that used parallel port now use USB.

Serial port-Many older digital cameras used this type of connectivity. Computers often only have one serial port and there is no way to connect devices in series. So, if you have an old serial device such as a Palm Pilot and a digital camera, you have to swap the connection frequently. This can lead to recognition problems (computer can’t detect your camera).

USB- This is the connection of choice for digital cameras and scanners at the moment. Very user friendly, a hub can be installed allowing over 100 USB enabled devices to be used on a computer with only one USB port. A device and be connected and used without having to reboot the computer. The next digital camera, printer, or scanner you buy will most likely connect via USB. A newer version USB,  version 2,  is said to be up to 40 times fast than USB v.1. USB v.2 devices are backward compatible with USB v.1 ports, but you don’t get any speed advantage. You have to have a USB v.2 device and port to get the higher speed of data transfer. A hair faster than Firewire 400.

SCSI-This used to be the most used connection for scanners. Still widely used with higher end scanners. The big problem with this connection is that you probably don’t have it on your computer. Generally, it requires the installation of a SCSI board into your computer. Most people shy away from taking their computer apart, hence it is not so popular. Not used by any consumer level digital cameras.

Firewire/I-link-IEEE1394-three different names for the same thing. Superfast and super small.  Currently used on some digital video cameras and some professional digital cameras. If you have an older computer, you may not have this connection.  If you don’t have a firewire port on your computer at the moment, you can buy a card for less than $50 that you can install yourself or have professionally installed. A new higher speed standard of firewire called Firewire 800 has recently been released that bests USB 2.0 for transfer speeds. Very few computers or other products use it yet.

CARD READERS

I strongly suggest you buy an appropriate card reader at the time you purchase your digital camera. If you think of your cameras memory card as a fancy floppy disk, then a card reader is just a fancy floppy disk drive. Generally, it connects to your USB port (faster, more expensive models are available that connect via Firewire). Once you install the driver (not necessary to install the driver if you use Windows XP or Visa), your card reader will be assigned a drive letter just as all your other drives have a letter assigned to them. You would then get files off the card just you get files from a floppy disk or CD. Card readers make living with your digital camera so easy. They are only around $20 for those that connect via USB. Just make sure that you get the appropriate card reader for the type of card you have. Many of the newer ones take up to 18 different kinds of cards. If you have a laptop, you might want to consider the purchase of a PC card adapter.  They are essentially a PC or PC Express card with a little slot into which you slide in you card.

 

FILE TYPES

TIFF- short for Tag Image File Format- This is the format that is most widely used for saving images in a lossless format that is also very cross platform friendly. It allows for a small degree of lossless compression using the LZW option given by some programs when saving a file. Since little or no compression is applied when using this format, the file size remains large.

JPEG or JPG- short for Joint Photographic Expert Group- while no longer the best format for compressing digital images since it degrades the image in proportion to the level of compression applied, it is the type to use for web applications and file sharing due to its wide spread use and acceptance.  When digital cameras compress images they almost always compress them as JPEG files because it is a sure bet that almost all computers out there will have the ability to open them.

PNG- A relatively newer file type, it offers very effective lossless compression. The problem is that it is not anywhere near universally supported. If you send a picture in this format to your friend, it is likely that they will not have the ability to open it without first installing some software.

PICT- An older, macintosh only format. Fine for Macs, but since it isn’t supported by Windows, you couldn’t share it with your PC loving friends.

BMP- an older type which is not appropriate for photos. It can’t handle all the color information that is necessary.

GIF- short for Graphic Interchange Format- not really appropriate for photographic images. It can only handle a very small color palette. Widely used on the web for simple graphics like billboards and welcome banners. Most animated banner ads you see on websites are animated gifs.

RAW FILE FORMATS

Many newer, higher end digital cameras are offering a new type of file format called a ‘raw’ file. These files are the actual image data as it is captured by the camera’s sensor. The data is not processed internally by the camera as all the other file types are. There is two major advantages of this. Most importantly is that the information is usually at a higher bit depth than the TIFF or JPEG would be saved in (see the explanation of Bit Depth above). This higher bit depth results in smoother gradations in tone and color. Also, no white balance or sharpening has been applied to the images saved in this file format. Thus, you can assign the color balance and degree of sharpening after having shot it. If you want the absolute highest image quality you can get, you should seek out a camera that has a raw capture mode. The major disadvantage of this file type is that you have to apply various settings afterwards using a propriety software package. A very rudimentary version of such software is generally included with the camera. A more featured version is often available for an additional fee. The Nikon raw file format is a .NEF (Nikon Electronic Format). Adobe Camera Raw, SilkyPix, UFRaw, Lightroom, Capture One, and Bibble are programs that allow you to process and edit raw files from a number of different manufacturers.

FOCUSING SYSTEM

All but the absolute cheapest digital cameras employ an automatic focusing system. Some of the more advanced cameras have multiple focusing points that you can use to focus on an area other than the center of the frame. Some cameras include a ‘manual focus’ feature that allows you to select a determined distance (usually by setting a numerical value in feet) to focus at should the cameras auto focus not meet your particular circumstance. Some simply let you set the camera to an ‘infinity’ setting that tells the camera to focus at it maximum focusing range which is generally something along the line of 30ft through infinity. Having an large range of distances that are in focus (great depth of field) is what disposable cameras rely on. Your cameras always want to focus on what is closest. If you try to take a picture through a screen door, you are likely to get a picture of a screen door. Same thing for trying to shoot through any kind of wire fence. In these circumstances, use your cameras manual focusing capabilities or set your camera to its ‘infinity’ setting.
Remember that you camera always focuses on a plane that is parallel to the back of the camera (the film plane). By changing the angle from which you are shooting you can determine what will and won’t be in the plane of focus and thus what will and will not be in focus.

FILM PLANE

The imaginary plane that extends outwards from your film or imaging sensor. This plane is of importance because at whatever distance you focus, there is a plane that is exactly parallel to the imaging sensor. If you hold your camera such that the film plane is parallel to that of a wall you are focusing on, than the entire wall would be in perfect focus. If you were to tilt the camera and thus change the orientation of the film plane without changing anything else, than only part of the wall would be in focus as the plane of focus would no longer be parallel to the film plane.

FLASH

Auto- camera determines when there is enough light present to properly expose at least part of what is in the frame. Works well for most situations.
No flash/flash off-flash will not fire regardless of lighting conditions. Use this when the object is beyond the reach of the flash.  For example, shooting a soccer game.
Forced flash–forcing the camera to flash regardless of the lighting conditions. If a person is strongly backlit, the flash will not fire in auto mode because it doesn’t know that you want to take a picture of the person and not the background. By forcing the flash to fire, the light level of the person in raised to more closely match that the background.
Fill-in flash-Same idea as forced flash. The flash fires to fill in darker areas that may be in the shadows due to back lighting.
TTL flash-the best kind. Stands for Through the Lens. It means that the flash is automatically stopped when a meter detects that enough light has come through the lens for proper exposure.
Slow Sync- slow synchronization of the flash with the shutter. This means that the flash will fire to illuminate the foreground subject, but a slow shutter speed will also be used to allow enough light in for a proper exposure of the background. For a perfectly sharp photo, a tripod should be used, but very interesting and pleasing effects can still be obtained without the use of one.


Experiment with using the flash during the day and no flash at night as discussed in class. Remember that taking bad pictures with a digital camera doesn’t cost you a cent so be prolific in your experimenting.
-use flash when something is strongly backlit
-only use flash when the subject is between 7 to 20 feet. Any closer and the flash will be too bright leading to bright highlights with no detail on the face. Any further and you are likely to exceed the flashes ability to reach that far.
-When taking picture with flash inside, turn on all available lights to reduce redeye
-Use the redeye reduction feature of your camera. You may still have redeye in the pictures, but it will be less than if you hadn’t used the feature at all
-You must use a tripod when taking pictures that require a long exposure time such as pictures of a city at night
-When taking picture of people, do not have them stand right against the wall. This leads to terrible shadows on the wall. Have them stand 5 feet or more from the wall when possible

Backlighting-When you take a picture of something where the main source of illumination (the sun, a table lamp, a bright window) is behind the subject, the camera makes its decision of how much light is needed with more weight generally being given to the very bright area. This leads to a photo with a well-exposed background, but grossly underexposed subject matter. This often results in a silhouette. Ways to avoid this problem: use forced/fill in flash to raise the brightness values too more closely match that of the background.  Avoid backlighting. Don't have the kids stand in front of the window for the prom photo.

No comments:

Post a Comment