Experiments with a miniature CCD
assessing a 'camera on a board' for amateur video microscopy
by Dave Walker
Click on the coloured text for 'jargon buster'
The price of CCD video cameras for security use is steadily falling and this is good news for amateurs wishing to venture into dedicated video microscopy. The cheapest cameras new are the miniature CCD 'cameras on a board' and prices in UK electronics catalogues range from £75 for a camera with a proprietary lens to about £100 for one with the standard 'C' lens mount for which most video microscopy adaptors are designed for.
I recently had the opportunity to try one of these covert cameras for video microscopy before it's use in a security application. On paper at least they seem to offer a potentially cheap route into basic video microscopy. The amateur at the moment is faced with the tricky decision of buying a camcorder, which is less than ideal for dedicated video microscopy work but has greater versatility in a domestic environment, or purchasing a security type camera either new or second-hand for dedicated work. There are pros and cons for going either route which have been well discussed elsewhere, but suffice to say amateurs have demonstrated that both are capable of good results.
Anyway, back to this dinky camera, an image of the one I tried is shown right. As you can see its barely larger than the 'C' lens mount and is shown next to a 35mm film cassette for comparison. Its specifications however are quite creditable. It's black and white with a stated resolution of 380 horizontal line pairs with a sensitivity of 1 lux (at f2.0). This resolution is comparable to a Hi-8 colour camcorder. It has a 1/3rd inch CCD sensor (500x582 pixels). It has no on board controls so any adjustment of contrast and brightness must be done on the monitor and thus wouldn't affect recorded images. (Many older security cameras available on the second-hand market are also like this unless an experienced video technician has internally adjusted the black level for video microscopy work).
The camera requires an external 12V less than 100mA power supply so I used a small off the shelf mains adaptor available from high street outlets. There is only one video signal wire out (the standard 1V p-p PAL composite video) that can directly feed UK video recorders and TV/monitors with an 'AV' input. The video earth is shared with the power input, so is easy to wire up as long as the polarity of the 'volts in' is carefully observed.
For the trial, putting the unit in an enclosure wasn't justified, although great care in touching the electronics to avoid shorting with damp hands or static damage was exercised. (Disclaimer: Don't adopt this approach with any purchased open circuit board electronic projects unless you are absolutely certain the equipment runs on very low volts / current via a properly isolated mains transformer. Even then modern circuits are easily prone to static discharge damage unless properly enclosed).
I should point out that I'm not an expert in video cameras or video microscopy. This is a very subjective quick test based on my experiences on using a variety of standard security type cameras for macro and micro work on an amateur basis. My own camera that I use for video microscopy is a Panasonic CL-350 with 430 horizontal line pair resolution and this served as the comparison for the test ('0.5 inch' sensor, 681x582 pixels, sensitivity 3 lux at f1.4).
When I first used the board camera I attached a 35mm SLR camera lens via an adaptor and used it for shots outside and inside. This gave a good contrasty and acceptably sharp image on my TV/monitor. In fact when I used my Panasonic in the same mode with the 'chroma gain' (colour) set to zero to effectively make it a black and white camera, it was hard to tell the difference between the images. I did notice that the board camera was very much more light sensitive as borne out by the specifications.
So far so good. I next tried it on the eyepiece of my stereo microscope (shown left above) using a 'C' mount to stereo eyepiece adaptor. In this role its lightness made it a potentially easy way of viewing e.g. pond life at low power on the monitor. The images were extremely lacking in contrast and very disappointing compared to those from my Panasonic camera when used in the same way.
I then used a standard eyepiece adaptor to mount it on my compound microscope. This is when I discovered that its greater sensitivity was a distinct disadvantage for the type of microscopy I did i.e. low power (40X to 400X) bright-field. Whereas my Panasonic was comfortable with a light intensity just a little lower than for visual use, I had to turn the lamp down to it's lowest level and still put every neutral density filter I could find in line to stop saturating the AGC (automatic gain control). This made quickly swapping from visual to video mode very tedious (the microscope doesn't have a trinocular head).
Although the images were very disappointing I did capture a few to my computer (captured with and without the use of a microscope eyepiece) to see what they looked like after tweaking. They were quite acceptable after heavy contrast improvement but hard work just to reach a standard my Panasonic was easily capable of without image enhancement.
So all in all I was disappointed in the performance of this small camera for video work, although admittedly I was comparing it with a colour video camera with many more features. I didn't have it long enough to identify if there were good reasons for its poor microscopy images cf. room images. I did check for flare bright spots in the optical train but I knew my set-up well enough to know that these weren't present.
Unless you do 'black cat in a coal cellar' microscopy i.e. high power dark field or phase contrast, from my limited experiences I suspect a very sensitive camera could be difficult to use for normal microscopy work. Especially if like me you use a microscope which doesn't have a trinocular head where in-line filters could selectively drop the light intensity in the video train.
Ironically the very small size of the camera was a distinct disadvantage for macro work with 35mm SLR lenses attached, as you would need to build an enclosure large and strong enough to take a standard camera tripod socket to support the video camera and lens. (I do a lot of macro work with 35mm lenses for digital still capture as well as true video microscopy without lenses attached.)
If you only have a £100 pounds or so to spend on a video camera I would tentatively suggest from this limited test, that miniature video cameras may not be the best buy. A second-hand ex-security camera with a larger sensor in a standard enclosure with tripod bush may be the better option (check it hasn't had a hard life - many have). Modern very sensitive cameras especially designed for low light use may also be tricky to use if bright-field microscopy was its main intended purpose.
Postscript: After publishing this article, I received some very valuable and pertinent comments from readers on the likely loss of contrast, which was probably related to infra-red sensitivity. Read a separate article discussing this, with some thoughts on the potential of infra-red video imaging for amateurs.
Comments to the author sent via our contacts page quoting page url plus : ('dwalker','')">Dave Walker
1) 'Aiming High - Camcorder Buying Guide' by Chris George. Article in 'Tecno' catalogue Winter 1997/'98.
Footnotes - an
attempt at jargon busting
(An article specifically discussing the features to look out for when buying a video camera for microscopy will be published in the near future.)
CCD - stands for charge coupled device. It is a solid state light sensitive sensor used in almost all modern domestic and professional video cameras including camcorders, security cameras and also digital still cameras. Older video cameras used vacuum tubes with light sensitive phosphors, e.g. Vidicon, Plumbicon designs. You may encounter this type if buying a security camera on the second-hand market, they are much bulkier and heavier and more prone to excessive light damage than CCD's. The larger tubes e.g. 1 inch would probably out perform a cheaper CCD in terms of line pair resolution so may be worth seeking out if high resolution amateur microscopy is of interest. CCD's cameras with a comparable resolution e.g. >500 lpm, are still very expensive, although the new digital camcorders are now achieving this. Return to article.
'C' lens mount - this is the standard lens mount for security style video cameras and most designed for video microscopy. It will accept so-called 'cine-lenses'. A cine lens is not required for video microscopy, the camera is fitted to the microscope with special adaptors that use this 'C' mount. Some adaptors use an eyepiece others don't. Special so-called relay lenses are also available for projecting the image into the video camera. Note that domestic camcorders don't use a 'C' mount and the lens is not normally removable, but amateurs have successfully used camcorders by focusing the lens at infinity, switching off auto-focus and projecting the image this way. Return to article.
Resolution - the resolution of a video camera is normally quoted in the number of horizontal line pairs (i.e. one black and one white) that it can resolve. To put things in perspective here's a few guidelines on typical resolutions (source ref. 1).
'Low band' domestic camcorder (e.g. VHS-C, 8mm)
- 230 line pairs
'High band' domestic camcorder (e.g. Hi-8) - 380 line pairs
New digital 'Mini DV' camcorders - 500+ line pairs
Domestic PAL VHS video tape recorder - 240 line pairs
Domestic PAL S-VHS video tape recorder - >400 line pairs
Cheaper CCD video 'security' cameras - 300-400 line pairs
More expensive CCD video 'security' cameras - 430-500+ line pairs
Return to article.
For video microscopy it is useful to have the black level (or pedestal) set so the blacks really are black. On older cameras this is usually an internal adjustment for an experienced video technician. On more sophisticated cameras setting this parameter is under the users control, on either a knob on the camera or a monitor screen menu. Return to article.
The sensitivity of a video camera to light is quoted as the minimum illumination in lux it can cope with at a given f number on an attached lens. The lower the value of the lux quoted the more sensitive it is. Return to article.
Please report any Web problems
or offer general comments to the Micscape Editor,
via the contact on current Micscape Index.
Micscape is the on-line monthly
magazine of the Microscopy UK web
site at Microscopy-UK