resolution

[Guest]

Obviously Georg has quoted the customers price and not the trade one.

[Guest]

That would be this camera here that Robert already linked to George.

Stangley they quoted £**k to £**k to me, is it coz u is Finnish?

53695[/snapback]

Might be that im Finnish but i was talking about

[UltraHighResolution]

Hello all

What a fascinating thread. Fathead alerted me to it so I thought I would contribute. Brief intro - I am Robin the SentryScope man (21 Megapixel).

Just to clear a few things up I'll start from the back.

Unless we were absolutely certain only Installed end user prices were given out at IFSEC. Currently UK customers include Police forces, Councils and Housing Associations. The end user installed price has been between £20-26,000. We only supply installation companies.

SentryScope systems are a one to one system 1 camera to 1 Server. Prices are for a complete system including the DVR so are roughly equivalent to 4 good quality dome cameras and a DVR.

A basic explanation of what we do - we are using a linescan camera 2100 pixels deep by 1 pixel wide, using a mirror we then scan 10,000 lines across to give us our 21 Million pixel image in a 90 degree field of view. Using both JPEG and MPEG like compressions we are able to reduce our original 21Mb image to between 1-5Mb and save on the SentryServer. Current processors/ Hard drive limitations mean that this process takes just over a second so our limitation is a maximum record rate of 50 images per minute. Even if we are zoomed into an image we are still recording the full 21MP. What this means is that we can zoom into a playback image 40-50 times and still have the same resolution as a CCD camera pointing in the right place at the right time with the right lens.

To put this into perspective with 6 SentryScopes we can cover a premiership football stadium and in playback identify every single face in the crowd, calculate the cost of simply cabling a conventional system to do this and SentryScope sudddenly becomes a very cheap option!!!

Regarding the comment from Rich "the video evidence when used in court has to meet a certain criteria, i.e. for identification the subject has to cover 120% of the recorded image size, not 120% of 10% of the original recorded image.

Also don't forget the other most important part, the lens." Actually in the advent of digital evidence this is no longer true. There is a Home office recommendation that for Facial identification they require 40 Pixels in 1 foot (80 for a number plate). Doing the maths it works out that SentryScope fulfills this criteria in an area 256ft wide by 163ft deep using an 85mm lens.

Hopefully this has cleared up a few of the issues mentioned in the thread, happy to answer any other queries here or direct to my email address, please let me know if you would like a demo disk, it really does help throw some light on Fatheads 'Daft ideas'

Robin@envisagetechnology.com

[Rich]

Hi Robin

Welcome to the forums, Thanks for making the requirements for digital cctv images clear.

Hope you enjoy your stay.

[Guest fathead]

Hello all

There is a Home office recommendation that for Facial identification they require 40 Pixels in 1 foot (80 for a number plate). Doing the maths it works out that SentryScope fulfills this criteria in an area 256ft wide by 163ft deep using an 85mm lens.

Robin@envisagetechnology.com

54746[/snapback]

Thanks for your post Robin, the 40 pixel per foot rule really puts things in perspective.

No doubt this is an absolute minimum, when you take video compression artefacts, noise etc into account I expect you need more than that from video in reality. Anyone who has tried still image capture, even from a DV cam, will know it bears little resemblance to the quality of image one can get from the most modest, now obsolete, digital still camera.

I note the Sentryscope does not compress the images inside the camera, I take it there is currently no compression chip that can handle the data fast enough. Is this something that may be in development for future Sentryscopes?

[UltraHighResolution]

"

I note the Sentryscope does not compress the images inside the camera, I take it there is currently no compression chip that can handle the data fast enough. Is this something that may be in development for future Sentryscopes?"

Sorry Fathead

I guess I oversimplified things (you learn to do that after a busy week at IFSEC). We use proprietry compression algorithms but without giving too much away we use a 'JPEGLike' compression in the camera and a further 'MPEGLike' compression in the DVR. We are able to do the camera compression because unlike conventional CCTV we are using a TDI Linescan camera and a mirror so we are compressing 'chunks of lines' whilst other parts of the image are still being captured. The SentryServer is Rebuilding the image and compressing further.

In the linescan world a 8400 x 96 Pixel camera has just been released so we are examining the possibilities of a 90 degree 84 Megapixel camera and perhaps larger angles.

[Guest fathead]

"

I guess I oversimplified things (you learn to do that after a busy week at IFSEC). We use proprietry compression algorithms but without giving too much away we use a 'JPEGLike' compression in the camera and a further 'MPEGLike' compression in the DVR. We are able to do the camera compression because unlike conventional CCTV we are using a TDI Linescan camera and a mirror so we are compressing 'chunks of lines' whilst other parts of the image are still being captured. The SentryServer is Rebuilding the image and compressing further.

In the linescan world a 8400 x 96 Pixel camera has just been released so we are examining the possibilities of a 90 degree 84 Megapixel camera and perhaps larger angles.

54899[/snapback]

Thanks for that

The mist has now cleared.

As for future plans for an 84 megapixel cam

Speechless

[Rich]

Speechless