| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Features Section |
Department of Orthodontics, Guy’s Hospital, London, UK
Address for correspondence: D. Bister, Department of Orthodontics, 22nd Floor, Guy’s and St Thomas’ Foundation Trust, Guy’s Hospital, London Bridge, London SE1 9RT, UK. Email: d.bister{at}doctors.org.uk
 |
Abstract |
|---|
 Top Abstract IntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
Key words: Clinical, photography, orthodontics, digital
|
Introduction |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
,2
‘Depth of field’ is defined as the zone of acceptable sharpness, and extends in front of and behind the point of focus. For good depth of field in intra-oral photography a small aperture (such as f=22) should be used. Consistent exposure at such settings can best be achieved by automatic ‘through the lens’ (TTL) metering.1 These settings should either be factory pre-set or should be easily programmable.
Although there are a large number of digital cameras advertised for use in dental imaging, only Single Lens Reflex (SLR) cameras consistently fulfil all the above parameters. Additionally, SLR cameras allow for ‘best preview’ of the proposed image as the viewfinder shows the object as it will appear in the eventual image.
In the pre-digital era the ‘Yashica Dental Eye’ was produced for intra- and extra-oral photography. This SLR camera had a 100 mm macro-lens, which was permanently fixed to the body with a built-in ring-flash. A macro-lens allows for distortion-free imaging at close range with high depth of field. The aperture settings were automatically adjusted when changing the focus from intra- to extra-oral photography, and was thereby very user-friendly; no changes in camera settings are necessary between intra- and extra-oral photography. It was also reasonably priced. The ‘Dental Eye’, however, will not be made available in the digital format. In this article, this camera was used as reference, as it had all the above-mentioned ideal properties for dental photography.1
Digital SLR cameras have only recently become more affordable. SLR cameras are now subdivided into two groups: professional and ‘prosumer’. The professional group is more expensive, but usually has a range of advantageous features over the ‘prosumer’ group such as:
However some of these features are also found in the ‘prosumer’ group; the Canon EOS 20 D is fast: 0.2-second response delay on turning the camera on and the Olympus E1 has sensor clean.
Although other authors2,3 have previously tested digital cameras, no comprehensive review has recently been undertaken. The professional group was not tested as most of the above-mentioned features are not necessary for dental photography.
For good illumination a macro-flash is ideal: it avoids shadows from cheek retractors at close range and allows for homogeneous illumination of the teeth. Ideally, these macro-flash units should be strong enough for extra-oral photography; thus, avoiding the need for a second flash system. With regards to the macro-lens and macro-flash the manufacturers’ respective devices were tested.
|
Aim |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
for the body), the so called ‘prosumer’ models (comparable to the ‘Yashica Dental Eye’) with manufacturers’ recommended macro-lens and flash in terms of:
|
Materials |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
) were single lens reflex (SLR), with a macro-flash and a macro-lens, as recommended by the manufacturer. The guide-number for the flash-units represent their power: the higher the number the more powerful the flash unit. One such assembled unit is shown in Figure 1. The Fujifilm S 3 Pro was tested with a Nikon SB 29S Speed-light and 60 mm Nikkor macro-lens.
|
|
|
Methods |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
). The camera was then used for extra-oral photography and the settings changed again until appropriate, in this case an aperture of at least f=8. The number of changes necessary between the settings was recorded. The camera was only considered metering ‘through the lens’ (TTL) if the flash settings did not need to be changed.
|
). A non-clinical method was preferred to taking images of a patient: the 10 cameras were not all available at the same time and changes in oral hygiene may have influenced colour consistency over time. Even if all cameras had been tested on one patient on one occasion the discomfort would have been considerable. The white balance selection was auto for all cameras. The images were subsequently downloaded on a ‘SONY VAIO’ (Sony Corporation, Japan) computer. The computer screen was adjusted to project the original computer generated colour using ‘SpyderPRO’ by ColourVisionTM (U.S.). The images were also ‘anonymized’ and assessed by two operators (F.M. and D.B.) for colour reproduction. Neither of the assessors were colour blind.
|
Two bodies were held simultaneously against the right and left eye of the examiner; the backdrop consisted of five fluorescent light tubes. The cameras were then swapped around to avoid eye dominance distorting the findings. Two operators (R.A. and D.B.) ranked the viewfinders for size and brightness independently, and the consistency was 100%.
Quality of the macro-flash
A manual stopwatch was used to measure the time needed to recharge the macro-flash between two exposures (precision 1/10th of a second). This was after the flash was initially charged and ‘tested’ with an empty exposure. Two measurements were taken and the mean calculated. It was also recorded whether the flash was powerful enough for extra-oral exposures. Each unit was tested with new single-use standard AA batteries (Duracell, NVSA Belgium).
|
Results |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
.
|
. All images were taken in JPEG mode.
User friendliness
Unfortunately, no camera produced satisfactory results in the factory default ‘automatic’ mode. This is because the manufacturers assume that the camera will be used for normal photography and the ‘factory default settings’ reflect this: the aperture will be programmed to be comparatively open in order to give the flash unit an increased range. Therefore, all cameras had to be adjusted at least once (initially) before taking satisfactory images.
Quality of the viewfinder: Inter-observer reproducibility
Quality and size of the viewfinders were scored with 100% consistency between the observers.
Quality of the macro-flash
Recharge times varied considerably and some of the units were true ‘ring-flash’ units (Canon), whilst others were more like ‘close range dual flash units’ (Nikon). However, there was no discernible difference regarding the homogeneity of the light.
Cameras
In the next section cameras will be discussed in alphabetical order according to manufacturer:
1. Canon EOS 350 D and EOS 20 D (Canon Inc., Tokyo, Japan)
The Canon EOS 350 D has 8 million pixels and has a plastic body. The camera has a small viewfinder (7th place), which is not particularly bright (6th). Handling characteristics are essentially identical to the EOS 20 D (Table 2).
The EOS 20 D, which has 8 million pixels, has a sturdy magnesium-alloy body. The size and brightness of the viewfinders were amongst the best (3rd and 2nd place, respectively).
After initially setting both cameras to aperture priority (f=22) a second adjustment had to be made to allow for adequate flash synchronization in this mode. Unfortunately, the flash synchronization mode is hidden in one of the sub-menus. However, once this was set up, only the aperture had to be adjusted between intra- and extra-oral views.
Both cameras were quite different in terms of colour reproducibility: the images of the canon EOS 20 D appeared slightly blue on teeth and gums, and the EOS 350 D slightly red on gums but blue on the teeth, when compared with the original model.
For both cameras the Canon MR-14 EX macro-flash was used, which had a comparatively slow recharge time (6 s) for the 100 mm Canon macro-lens.
2. Fujifilm S 3 Pro (Fujifilm Corp., Tokyo, Japan)
This camera is relatively new and has a variety of features, which are different from all the other cameras tested. It has ‘12 million’ pixels, of which half are dedicated for situations with little light and the other half are dedicated for situations with normal and bright light, hence increasing the ‘dynamic range’ of the sensor. However, for all but the ‘maximum dynamic range’ settings only half of the pixels are used, resulting in 6 million effective pixels. It allows the user to take images in a variety of modes: ‘maximum dynamic range’, ‘normal’, ‘film simulation 1’ and ‘film simulation 2’. Images in all modes were taken and individually scored for colour reproducibility. The best images were taken in ‘maximum dynamic range’ and these were subsequently used.
In contrast to its predecessors (Fujifilm S 1 Pro and S 2 Pro), this camera works in manual mode only when using a macro-speed-light and the settings of the flash have to be changed between intra- and extra-oral views. The camera is therefore not TTL when using the SB29S as flash unit (Nikon Corp., Tokyo, Japan). The camera does, however, allow true TTL metering when combined with the Sigma® EM-140 DG iTTL ring-flash (Sigma Corp., Kanagawa, Japan). The Fujifilm S 3 Pro and the Nikon D100 are based on the Nikon F80 body; the manufacturers specifications for the viewfinder are virtually identical and were therefore ranked identical for the viewfinder quality.
Colour reproducibility was very close to the original model, the images appeared slightly redder on the gums, and the teeth had a slightly blue tinge.
The camera was tested with a 60 mm Nikkor macro-lens.
3. Konica-Minolta 7D (Konica-Minolta Holdings Inc., Tokyo, Japan)
The exposures of this camera were inconsistent when taking intra- and extra-oral photographs in terms of lighting. The inconsistency was caused by the flash, which allowed exposures before full recharge. For assessment of quality of photographs only correctly exposed images were allowed.
The colour of the images appeared to put more weight on reds than the model, all colours were stronger than the original object.
Handling characteristics were very complex: in manufacturers auto-setting (P) the aperture did not allow for adequate depth of field. When changing to aperture priority and f=22, the macro-flash had to be separately adjusted as otherwise underexposure occurred. When changing to extra-oral photography both camera and flash-settings had to be readjusted. The viewfinder was the second largest, but brightest of all cameras tested. The Konica-Minolta R 1200 Macro flash was the second slowest to recharge, taking 8 seconds.
4. Nikon D100, D70s and D50 (Nikon Corp. Tokyo, Japan)
All three cameras were similar regarding their handling characteristics: they work in manual mode only when using a speed-light and both settings (camera and flash) had to be adjusted between intra- and extra-oral views. The cameras therefore do not use TTL metering when using a macro-speed-light. All cameras have approximately 6 million pixels.
The D100 had one of the smallest viewfinders, which was also not particularly bright. The D100 images were consistently scored closest to the original model amongst all cameras, tending slightly towards orange.
The D70 images were slightly lighter than the D100 images, but equally acceptable.
The D50 has the smallest and darkest viewfinder of all tested cameras and the D70s had the second smallest and dimmest viewfinder. The D50 images appeared to put more emphasis on the red and blue colours, thereby appearing more vivid for teeth and gums.
All three cameras were tested with a 60 mm Nikkor macro-lens and a Nikon SB29S macro-speed-light. The latter had the fastest recharging flash in the series (3 seconds).
Interestingly, all three cameras worked well when using the Sigma® EM-140 DG iTTL ring – flash. However, this was not scored as there are too many camera/lens/macro-flash permutations between manufacturers available.
5. Olympus E1 and E300 (Olympus Corp., Tokyo. Japan)
The two Olympus cameras are virtually identical in their handling characteristics. Intra-oral photography at a magnification of 1:2 on P setting gave consistently overexposed images. However, once the camera was adjusted to aperture priority and f=22, both cameras took good images regarding the exposure and depth of field. No changes were necessary when changing between intra- and extra-oral photography. In other words the Olympus Ring-Flash was powerful enough to take extra-oral images at an aperture of f=22. Also, the camera turns the flash unit automatically on and off. The Olympus E1 is a professional camera with a magnesium alloy body, is environmentally sealed and is a 5 million pixel camera. The viewfinder is somewhat bigger and brighter than the one of the E300, but both were in mid range compared with the other cameras. The E300 has a less rugged design and has 8 million pixels.
The E1 took slightly darker images than the E300, but both produced slightly bluer images than the original.
The flash used was the SRF-11 and the lens was a 50 mm ED macro-lens, both Olympus. The flash recharge time was 4 seconds (medium range).
6. Pentax *istDs (Pentax Corp., Tokyo, Japan)
This camera uses the same sensor as the Nikon D100. However, in contrast to the latter camera, the Pentax images were strongly biased towards red. The camera was also very ‘trigger happy’ causing multiple unwanted exposures. The intra-oral exposures were good once the initial change from factory preset to manual (f=22) had taken place. However, taking extra-oral views required changes to the aperture settings. The viewfinder was the largest of all cameras tested, but only mid-range when it came to brightness. The camera is comparatively small and was the lightest tested with 1324 g (including macro-lens and flash unit).
The camera was tested with a Pentax 100 mm macro-lens and Pentax AF 140C flash. The latter proved not very powerful, so great care has to be taken not to move too far from the patient for extra-oral views. Also the recharge time was the slowest at 9 seconds.
|
Discussion |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
Ease of use is particularly important for auxiliary staff, who may have had little previous training for dental photography. Only the two Olympus cameras matched the above benchmark parameters after alterations of the initial factory settings were made. All other cameras needed a change of aperture settings between intra- and extra-oral photography. It is interesting to note that instead of changing the aperture from f=22 to f=8 between intra- and extra-oral images, one could also change between modes: from aperture priority to ‘manual’ or ‘portrait’ mode as these modes often automatically change the aperture. However, this was not consistent between manufacturers. As changes were necessary for all models (apart from Olympus) changing the aperture is recommended.
Colour reproducibility
Measuring colour is complex—there are several systems available to aid this process. To our knowledge only one has been published regarding the impact of colour fidelity in orthodontics.4 In analogue photography, colour fidelity was largely determined by the make of film, but the flash unit, as well as the lens used also contributed. In digital photography mainly the sensor and the subsequent processing of the data determine the colour of the image. The difference in colour reproduction for digital cameras reflects the manufacturers assumed customer preference for colour weighting. The Fujifilm S 3 Pro exemplifies this best: the camera has two ‘film simulation modes’, which allow the operator to choose the weighting of colours. Generally, colour profiling cameras is often only achievable for specialist equipment: even professional 35 mm SLR cameras edit or ‘colour render’ images (ISO22028-1), thereby interpreting/assuming the characteristics of a pleasing appearance of the image. The colour rendering will, among other things depend on the format they are taken in (by example JPEG or TIFF): the camera acts as a computer ‘translating’ the images on the sensor into the respective format. In most ‘prosumer’ and all professional cameras the user is able to bypass this process by downloading ‘RAW’ (unprocessed) images to the computer. However, even when taking images in ‘RAW’ mode some rendering will take place in the software used to produce the images on a screen or printout. To undo the ‘rendering’ would therefore remove the images’ intended characteristics. This investigation therefore abstained from ranking the colour reproducibility of the images and only used descriptive terms to characterize the colour reproduction (for further information, see International Colour Consortium white paper5). Colour adjustments will have to be made for all cameras to produce images matching the original object for colour temperature.
In addition, for ideal reproducibility of colour on monitor and printer, an elaborate calibration process is required. Monitors need to be regularly re-adjusted as colour may change over time. For printers the colour fidelity may vary for different print-media and may also vary with each new print cartridge. In this investigation, ‘SpyderPRO’ by ColourVisionTM, a combination of hard- and software was used, which allows the monitor to express the original colour generated by the computer, to a high standard. However, even after monitor and/or printer have been calibrated, in the end, direct comparison between the object itself and its image will be necessary. Although colour temperature can be measured, the figures themselves should be treated with caution, since a difference in the measurement does not necessarily reflect an equivalent change in the observer’s perception. This final comparison between the object and the image will depend on a variety of factors: the temperature of the ambient light, temperature and humidity of air etc. It is also worth remembering that about 7% of male and 1% of the female population in the UK are colour-deficient.6 Slight variations in colour, although occasionally important, are not absolutely paramount to taking good orthodontic records. The main aim was to test for ease of use and hence alteration of images was not allowed.
It was often difficult for the assessors to consistently identify the images and from this it was concluded that there was little difference between the cameras.
Quality of the viewfinder
Testing the quality of the viewfinder requires all cameras to be lined up and compared by holding two cameras simultaneously against right and left eye. To our knowledge, there is no other objective method to test this feature. Brighter and larger viewfinders allow easier focusing and handling, and it was surprising to see that size and brightness of the viewfinder did not necessarily correlate with the price of the camera.
Time needed to recharge the macro-flash
The fastest units were re-activated within a few seconds. Although 6–9 seconds does not appear to be a long time, to us, it is long enough to cause delay in workflow. The units were tested with previously unused, standard non-reusable AA batteries (Duracell), and flash-recharge times may vary when different batteries are used.
It should be noted that most macro-flash units were not particularly powerful and great care needed to be taken not to move too far from the patient for the extra-oral views, as underexposure may result.
|
Conclusions |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
The cameras, which were consistently ranked best for colour-fidelity, were among the most complex to use (Nikon D100 and Fujifilm S 3 Pro). For ideal colour reproduction images of all manufacturers have to be adjusted, so this parameter is not to be the most important one when it comes to choosing a camera. Other factors, such as robustness, environmental sealing (water and dust protection—Olympus E1), weight, viewfinder qualities and ability to clean the sensor from dust at start-up, may all impact on the final decision-making process. Finally, there is a significant difference in price and this will undoubtedly play a role in the decision-making process.
Models are being replaced at a rapid pace, and industry may eventually develop a digital SLR dedicated for dental use, which is easy to use with the standard settings, such as the ‘Yashica Dental Eye’.
|
Notes |
|---|
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionAimMaterialsMethodsResultsDiscussionConclusionsReferences |
|---|
2 Sandler J, Murray A. Digital photography in orthodontics. J Orthod 2001; 28: 197–201.
3 Doldo T, Fiorelli G, Patane B. A comparison of three digital cameras for intra-oral photography. J Clin Orthod 1999; 33: 588–93.[Medline]
4 Eliades T, Kakaboura A, Eliades G, Bradley TG. Enamel colour alterations associated with orthodontics. In: Graber TM, Eliades T, Athanasiou AE (Eds) Risk Management in Orthodontics: experts’ guide to malpractice. Berlin: Quintessence Publishing, 2004; 11–18.
5 International Color Consortium. White paper #17. Available at: http://www.color.org/ICC_white_paper_17_ICC_profiles_with_camera_images.pdf (accessed 30th May 2006).
6 Cumberland P, Rahi JS, Peckham CS. Impact of congenital colour vision deficiency on education and unintentional injuries: findings from the 1958 British birth cohort. BMJ 2004; 329: 1074–5.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
