Cooperman

My initial impressions are that the imaging problem appears to be more to do with lighting than anything else (unless bosch cameras prefer BMW's to Peugeots ) Where a vehicle has it's fog lights on, they're throwing light straight down to the ground, and the reflectance off the snow is lighting the front of the vehicle and number plate, thereby offsetting the iris closing effect caused by intense headlights. The rear view looks even better exposed, again because there are no intense headlights causing the iris to over compensate. Ideally, if the cameras fixed shutter can be used at 1/250th to reduce movement blur, and the ambient light increased onto the target area (visible light for colour operation, infra red only if the camera is photocell switched into B/W at night), then you should get a very reasonable result. Using a floodlight at that distance is not really going to be much use, as the lighting ratio between the highlights (the vehicle headlights) and the shadow areas will be outside the range of the imager. In other words, there just isn't enough light output from your flood, to offset the negative effect of headlights. Perhaps as an experiment try using either a much narrower beam light source(s), or take the flood much nearer to the target area (perhaps 10 metres away). Applying the "sens up" will not help, as frame integration will certainly increase sensitivity but at the expense of increasing the blur effect as the vehicle moves through the image. Incidentally, raising the height of the camera (although less than ideal from an operational point of view) does help to reduce the negative effects of vehicle lights. As far as I am aware, the ANPR cameras used for the London Congestion Charge scheme, are all mounted well off the ground, and optimised for plate recognition at perhaps 20 - 30 metres from the pole.

You could set up the filter rotation on the ground, tape them togethor, then that's one less thing to worry about when you're hanging off a ladder (been there, done that! Actually did it last week with a temporary two camera install). The polarisers should give a very dark effect at maximum (90 degree) rotation. If it only looks slightly grey, then there is something not quite right with the filters. If a filter scrathed easily, then it's probably poor quality plastic. Glass is slightly more expensive, but you get what you pa ..... etc. etc. Incidentally, depending on what cameras you are using, it's often easier to remove the camera to ground level to set up focus / exposure controls, and then refit it and align after you've done the main adjustments. If necessary you can use extension leads to reach the monitor at ground level, which is generally much safer than trying to use two hands for a four hand job.

Well said larry, and thanks again for providing this forum. Have a great one!

It's that time of the year again, so before I forget, a quick message for all forum members .... Enjoy your festive season wherever you are, and my very best wishes to you all for 2006! regards, Coops.

Somethings not quite right here rory, So how exactly did you use the polarising filters? Ignore what was written on the filter description page, you're not simply polarising the light in just one plane as described; you should be using both filters - the first filter polarises the light, and then when the second is rotated, it progressively blocks the rays from passing through, hence the variable ND effect. If sunlight is causing flare on the filter, then you should also consider buying a rubber screw on lens hood. I'd suggest you have a play around with the two filters just indoors for now - don't even bother with a camera. If they're not going dark as you rotate them, then they're not polarising filters. BTW $ 17 for a filter actually sounds very reasonable. Incidentally, there's nothing to say you can't filter out the light with either ND's or polarisers, and then also add the dichroic IR filter to focus at maximum aperture at a specific fixed IR frequency. If you focus shift the lens slightly off optimum for colour (low light) use, when the B/W kicks in, the increased depth of field resulting from the aperture closing slightly, should take care of your focus problem. During daylight, the small apertures (giving a larger depth of field) should provide a crisp image whether it is under visible or IR frequencies.

BTW, polarising filters look virtually clear, until you rotate one over the other.

You will need quite a dense ND filter combination to force the camera into monochrome. If you try two polarising filters (expensive!), hold one on top of the other and look at a light source. Then rotate one and watch it get darker. Individually, they're also great for cutting out windshield reflections under controlled lighting conditions, for example in underground car parks.

Try using your 840nM dichroic filter. It's the same frequency of light day or night, so it's only the light level which will vary after dark.

If it looks like a mirror, it's probably a dichroic filter with a coating bonded onto the glass substrate. You can hold the filter right up to the lens of a b/w camera, and it should see straight through it if viewing under daylight or tungsten light sources. As long as you don't spend ages staring at the filter with the lamp switched on, you shouldn't have any difficulties. You're far more likely to have problems coping with the bright sunshine in your tropical paradise

What did you see .....?

rory, you could always leave a stocking at the end of your bed on Christmas Eve

The filters don't have to fit ... you can simply hold them over the front (or tape them in place), while you make any adjustments. Usually 52mm diameter or larger are ideal for this type of work.

If the filter is 'dyed in the mass' glass, then there's no problem handling it. If it's a dichroic filter, then the coating on the glass may be easily damaged. With both types, you should never look through the filter at a bright light source for any length of time. Although the human eye is not capable of imaging with Infra Red at this frequency, prolonged exposure could damge the retina, and it's not reversible

Neutral Density Filters are normally sold by photographic dealers in graded steps, so for example NDx2, NDx4, NDx8 (x2 is equivalent to one 'f' stop on a lens). For focus setting in very bright light, you normally need a very dense level of ND filtering, so (for example) with a set of 3 filters, you can combine them to create anything from NDx2 > NDx14. With two polarising filters, you simply rotate one over the other to create an infinitely variable level of filter density. They work in different ways, but achieve similar effects, albeit there is a price differential, so you pays yer money and takes yer choice. Personally I prefer the flexibility of using two polar filters, but each to their own.

Ideally, setting a back focus in the least amount of usable light is the perfect situation, then you don't have to worry about ND filtering, you just have to put up with on screen noise the size of golf balls Most manufacturers back focus their cameras before delivery, so it's usually only an issue with odd or exotic lenses. The other way of approaching the subject is where (for example) you want to focus on an object about 30 feet (10 metres) away. Twist the lens focus ring slightly away from infinity (if it's marked up, then obviously leave it set to a point approximating 30 feet), then use the back focus adjuster to move the imager until the image is in sharp focus. If the camera is being used for a fixed view indoors, then this "alternative" method works just as well. Setting an iris pot, is light level (not location) dependent. If you have a very well lit room, adjust the pot slowly from maximum aperture until the image is correctly exposed. Then turn the pot perhaps a further 5 > 10 degrees, and that will still maintain correct operation, but allow for a small amount of tolerance under varying circuit temperatures. Personally, I always like to set cameras up when they're running at normal operating temperatures (if you set it up 60 seconds after power up from cold, it may well drift a few hours later, particularly if it's being used in a very warm environment). Also, if the camera is eventually being installed near a ceiling, the temperature can rise quite dramatically above that found on a test bench, and that can cause circuits to drift. If you shine a torch directly down the lens (or point the camera at the ceiling and move it slowly until the light source drifts into view), the iris should then shut down correctly, and reopen again as the light moves out of the field of view. No matter how well you set up a camera on a bench / in a lab / on site, you still have to double check that it's operating correctly when it's finally installed 'in situ'.

Congrats on the article rory At this rate you'll soon be a global surveillance celebrity (then who's going to do your cameras for you?). The only thing that spoilt it for me was the article above yours about CCTV on the London Transport network - if I see another quote from Londons Mayor saying how brilliant they've been at installing total c**p, I'm going to ...... Sorry, rant over ....

Absolutely rory! When you're using a zoom or varifocal lens, if you try and set the back focus at wide angle, the depth of field is much greater than at the telephoto setting, so it's significantly more difficult to find the optimum 'point of focus'. If you set the back focus at telephoto (with the lens focussed at infinity and pointing at a distant object), when you change to a wide angle setting the depth of field increases, so there is a better than reasonable chance that with a quality lens, the image will remain sharp throughout the zoom range. You don't need a correctly exposed image to focus a lens, just so long as you can see the shadow components, it doesn't matter if the highlights white out completely (although excessive flaring with cheap optics can make the job even more difficult). To be honest, it's very easy to get hung up on the 'back focus' issue, where in reality, it's rarely as crucial as we might at first think. If a box camera is going to be fixed in position fitted with a single (non motorised) lens, once that image is correctly focussed it's unlikely to be changed unless the camera is subsequently redployed elsewhere. As long as the image is correctly focussed, that is really the most important thing. From a professional perspective, I always wanted my cameras to be set up correctly, so that if any subsequent installers decide to move it, or fit alternative lenses, it will work first time for them, and they won't be cursing the useless individual who couldn't even get the back focus set correctly It takes less than 60 seconds to set a back focus, so I don't begrudge spending the time to get it right. It's worth remembering, you can always set the back focus using a preferred 'benchmark' lens, preferably a very high quality fixed focal length optic (perhaps a 16 /25 / 50mm), and then any other quality lens can be fitted and it should work correctly without any further readjustment. If you use a manual iris lens (set to maximum aperture) with the cameras 'Electronic Iris' function switched on, then adjusting the correct back focus is a piece of cake. One other quick point; with very cheap and nasty lenses, the back focal point can even vary between lenses of the same batch!

Thanks for the link CCTVmofo; I had a quick look and there are a few things there which are, shall we say, less than accurate, but I won't hold that against them. IMHO most of the information is very good (but not all). Strewth rory, a 13" TV up a 12' ladder!! I used to use a 9" B/W monitor myself, because we didn't have any small test monitors in those days. One thing you could try - I'm assuming you won't find it easy to get hold of an anti glare filter (you can get some reasonable ones for computer VDU's), maybe check out whether you can get a lightweight slatted wooden louvre. You only need to see small patches of screen to correctly focus, and picture quality always looks different when you see it on a monitor in a darkened room, but it might be worth a try. Incidentally, when focussing cameras (during the day) being used with tungsten IR lamps, I used to remove the filter part of the illuminator, and hold that over the front of the lens to allow re-focussing at the correct light frequency. 850nM is the same whether it's filtered from the sun or a tungsten bulb. Of course the quantity of light will differ, and you have to realign illuminators (at night) for correct night operation, but usually the focussing can be done during the day. Now admittedly, I have a set of IR filters between 800 and 1000nM, but they rarely get used these days. Take care up that ladder rory!

tmartin, The issue of playing back or copying recordings for evidential purposes, is not simply a technical issue relating to compatible equipment. Whilst copying a recording onto a portable device may seem like a straightforward solution, there are potentially serious quality loss issues that need to be considered. Also in some countries, evidence submitted to court has actually been thrown out for various reasons related to the original source material (a very high profile court case in Australia, the Falconio murder trial, has only just ended despite the CCTV recordings being thrown out by the judge). It would be worth researching some of your law enforcement colleague sites, that may have some detailed and useful information relating to recorded evidence in the States (our rules over here in the U.K. are quite different ) You could maybe try the LAPD site, SWGIT, and a few others that I can't think of off the top of my head. Only after you have a better idea about the L.E. perspective on handling recorded evidence, should you then consider what equipment you are going to need. As has already been suggested, it certainly wouldn't do any harm at this stage to make enquiries with both local supply companies and 'end users', to find out what the most common record systems are presently being used.

Sunny tropical conditions ..... in the U.K.? ..... Seriously rory, we do get fairly intense sun mid summer (contrary to popular myths), and my favoured option for adjusting cameras is to use two polarising filters, one on top of the other, held over the front of the lens. As you rotate one filter, you can vary the degree of "Neutral Density" filtering from absolute minimal to almost total blackout. Of course it's not really a neutral density set up, but it has exactly the same effect, and under any given lighting conditions, you can infinitely vary the filtration to achieve a correctly exposed image.

Hi Tarek, Four quick questions for you:- What are they stealing, where are they stealing it from, do you have an idea who might be doing the stealing, and ..... how often is it happening? Before you can address the problem with hidden cameras, you need to look carefully at the nature of the problem, otherwise you just risk wasting money, and possibly making the situation even worse.

Hi HECTOR, There's not a lot wrong with your english, It's a heck of a lot better than my peruvian (or should that be spanish?). The term DSP means Digital Signal Processing, so it actually doesn't refer to what is coming out of the camera. The BNC output is conventional analogue, but the DSP bit controls the internal workings of the camera, in much the same way as the engine management system on a vehicle optimises performance. There are lots of cameras on the market that benefit from DSP, but they are still essentially analogue CCTV cameras if they have a 1v p-p composite video output. As regards TVL and pixel counts, there are other threads on the forum that relate to this, and I'm sure someone will jump in here with a couple of links. Just for the sake of sanity in the workplace, try and stick to using just one relative measurement in the short term, perhaps Horizontal Resolution (e.g. 380 TVL, 420 TVL, 480 TVL etc. ) until you get the hang of what it does (or often doesn't) mean in the real world. If you try and relate pixel counts and bandwidth to actual images, you'll probably just end up going quietly crazy!

U2 .....? Aren't they a dublin based CCTV installation company CCTV_Australia, long time no see - how ya doin??

Classifying camera resolutions shouldn't be that difficult, but then not many things in life were ever meant to be easy. Any respectable manufacturer quoting 'x' lines resolution, would generally be accurate, provided of course that in the case of "high resolution" cameras, the lens fitted was actually capable of resolving to that degree. It is frighteningly common to find 480+ lines resolution cameras, fitted with poor quality lenses that effectively pull the performance back to standard resolution, or worse. For any given resolution camera, for example 480 lines, if a comparison of imager pixels is to be made, it is important to remember that the actual picture element count will vary with EIAJ / NTSC against CCIR / PAL. That said, I have to agree with most of what has already been posted in this thread.

I still haven't found, what I'm looking for .....