dvarapala

^ this. BTW, for a turnkey (well, as turnkey as it gets) Asterisk installation, I highly recommend Trixbox. I've been using it for almost 3 years now, and I'll be adding the Mobotix doorbell cam to my system as soon as I can scrape up a spare $1000.

IINM, the M12's lens modules are interchangeable, as well.

Speed is typically 15 MPH or less. I'm at the end of a cul-de-sac, so cars have to slow down somewhat in preparation for turning around. Distance is roughly 30 feet. Also, a straight-on shot is not possible; the camera will be offset horizontally by about 15 - 20 degrees. One of my big concerns was the manual iris lens (and camera) being used in an outdoor environment. While there will be a visible light filter and full-time IR lighting, I can only imagine that there will still be a lot of variation in the amount of light. Does this approach even have a chance of working, or should I forget about the manual iris stuff and go auto?

Forgive me if I've used the terminology incorrectly. I assumed the "Recognition" part was simply the Optical Character Recognition algorithms required to pull the plate numbers out of the video frames. BTW, I'm not the only one who uses the term in this manner. The open source javaanpr package also lacks a database look-up component, and focuses solely on the character recognition: http://javaanpr.sourceforge.net/

Well, that works out just fine, since I'm a hobbyist and not a business. For that price, it had better be six-sigma reliable, and when it does fail, their lead software engineer had better be over at my place in a hour to fix it.

I'm working on a homebrew license plate capture system, as well. Any chance you could post a couple of captured frames from your setup?

Hammond makes some very nice NEMA 4X enclosures: http://www.hammfg.com/pdf/cat/Minicat_NM.ZIP

http://www.zoneminder.com Well, ZoneMinder is free, and so is manual review of the captured footage. If/when I go to automated capture, I'll be developing the software myself as a ZoneMinder plug-in. There is at least one open source ANPR package I can use as a starting point. But all that is beside the point; regardless of what kind of camera is feeding the images into the system, the back end hardware and software (and their costs) are going to be the same.

I've looked at a lot of specs for CCTV cameras recently and something that is almost never mentioned is the filter size. Is this because most CCTV camera lenses do not accept screw-on filters? Or is it because there is a standard size that they all have so nobody bothers to mention it? In the latter case, what is that size? Thanks!

Duct tape? That's a good idea, and I may end up resorting to that.

I'm still struggling with this. It seems that I not only have to worry about the diameter of the filter, but the thread pitch as well. For example, my Sony camcorder has a 37mm diameter and a 0.75mm thread pitch. Computar (as an example) megapixel CCTV lenses all seem to have a 0.5mm thread pitch with various diameters. I have been unsuccessful finding filters or even step-up adapter rings in this size and with these thread pitches. Perhaps I'm just not looking in the right places?

Here is the paper delivery (not for us) this morning probably about 15mph (cropped screen shot) off Vitamin D. I am pretty much zoomed out on the 3.3-12mm, set at 1/30s. Is your scene completely dark maybe? There are two streetlights, one on either side just outside of the frame. However, they are up on 40-foot poles; since your coach lights are closer to the ground, they may give you more usable light. True, although too much light can also be a problem. I have a flood light with two 90-watt halogen bulbs mounted on the wall near the camera; when it's turned on, the P3344 "hunts" between day and night modes every few seconds. Highly annoying. I'm now educating myself on IR illuminators.

Well I sure do. This is a scene without motion: This is the same scene with a car going by: This is with the default settings. If I set the camera to favor motion with a minimum shutter speed of 1/30 second, the picture gets a lot noisier with no real reduction in the motion blur.

I have to concur with this. I have a P3344VE and it delivers clear pictures at night as long as nothing is moving. If a car drives by, the motion blur is so bad that you can't even tell what kind of car it is, even though I'm at the end of a cul-de-sac and the cars are never moving more than 10 - 15 MPH tops as they pass by.

Naturally the adapter you need will depend on which camera you have, but here's where I got mine: http://webcaddy.com.au/astro/ProductList.asp?cat=adapter

If you want to keep the camera safely out of harm's way, it's possible to get phenomenal distance without spending $10,000. I have a small telescope (Celestron Nexstar 5). It's a 5" Schmidt-Cassegrain unit with an effective focal length of 1347mm. I bought an adapter which fits onto a $99 Logitech webcam and lets it slip into the telescope's visual back. The setup is intended for astrophotography, but I have used it quite successfully for terrestrial viewing; I have made videos of activity 20+ miles away using this setup. The telescope cost me about $1000, the camera was < $100, and the adapter was maybe $20. The downside is it's not PTZ.

That first one ("independent of my PC now") is the killer. That's pretty much going to limit you to cameras that will record to a built in SD card or other nonvolatile storage (e.g. Axis cameras). I suspect you'll be much happier if you take an old upgrade orphan PC (don't we all have at least one?) out of the closet and put it to work as a DVR/NVR.

Wow, that's amazing! What kind of an image do you get from that M12 in the daytime?

$300 would probably include his markup. Doesn't the MSRP already cover that?

Dunno if you're still looking for switches, but I have been using a Trendnet TPE-80WS 8-port (all gigabit) PoE switch for several months now. Street price is ~$250. It does have a small fan in back, but it's been sitting under my desk for almost a year and I have never heard it.

This information comes straight from the horse's mouth:

I'm planning to install a dome cam on this solid wood post, in roughly the spot labeled '1': This post supports a solid wooden beam that runs the entire width of the house. My current plan is to run some gray PVC conduit along the top of this beam to reach the post: The whole area is sheltered by the roof, but a strong wind might blow some rain onto the camera and conduit; however, the PVC conduit should be reasonably water-tight so I should be all right, shouldn't I? The area labeled '2' is the outside wall of the garage; this would be the easiest wall to penetrate to bring the wires into the house. At the post end of the conduit, however, I'm having trouble coming up with a good way to make the right-angle turn to run down the post to where the camera will mount. Any ideas on how to complete the run? Is there another approach I should consider? Thanks for sharing your wisdom.

Thanks for the ideas, Soundy. Hmm... I thought I read somewhere that ENT was not intended/suitable for outdoor use?

This is all fascinating, but there are far too many variables at play to make any valid comparisons. Scenes, lighting conditions, shutter speeds, CCD sensors, lenses, how well the compression algorithm is implemented, whether hardware acceleration is used or not, and any number of other factors may cause the differences you see in the images. The only way to scientifically compare H.264 compression to JPEG compression is to start out with the same (uncompressed) image, run it through the compression algorithms, and compare the outputs. That's the only way to control all the other variables except the one you wish to measure.

I was about to suggest we all sign up and weigh in It may be too late for that; it looks like the thread had been deleted. Was this the one where the professional security consultant bid $100 to install some sort of networking equipment at the end of a 900' underground cat5 run only to discover that Ethernet's maximum distance limitation is about 1/3 of that distance? So his solution was to twist a couple of the unused pairs together and run 120VAC down to the other end so he could power a WiFi access point?