NotoriousBRK

No problem... It was kinda crowded anyway, but I wish you could have hung out more. I think you left around 1:30, we were only there for a little over 2 more hours

I didn't do any major events this year either. Monday, Tuesday, Wednesday nights was generally customer dinners and business meetings. I think I was back to my hotel by 1AM easy those nights. Thursday I threw together a last-minute thing at Tao just because I didn't want to leave without having had a little fun. We closed the place down, including an appearance by Coolio at about 2:30 AM in the back room where I had a table.

Yes, our system should be able to provide the ability to distinguish between a person and a deer (or other animal) in most cases. VideoIQ does not use "motion detection" analytics, and it does not require you to spend all kinds of time tweaking and tuning the system in place. Basically, we use a "pattern matching" system, where the system learns the background of the scene and compares the speed, shape, size, color, texture map, movement direction/style, and other parameters of moving objects to an on-board database of over 200,000 visual patterns and descriptions that have been learned into its algorithms. Based on this it is able to auto-calibrate to the specific scene/installation where it is installed. ANY video analytics system is going to be dependent on the overall visual quality of the scene, meaning that things like good focus, lighting and lack of distortion in the image will contribute directly to the quality of the decisions it makes. Or, to put it another way, coupling a crappy camera to an advanced system will tend to yield crappy output You can take our encoder and attach it to a good quality analog camera, or you can use our iCVR-HD megapixel camera and get really good images. The HD camera is the most popular choice, and we're currently the only company doing real-time megapixel analytics. To address your specific question though, there MAY be times where it is difficult for the system to distinguish between a human and a deer. A deer walking directly towards the camera is going to have a lot of the same visual characteristics as a human. It is going to be about the same height and aspect ratio, have similar movements, etc. There are things you can do with our system to manage this to a large degree. Rules have a detection sensitivity setting, which allows you to specify how good of a match an object needs to be in order to trigger the rule. You can also set parameters like a detection time. Basically a loitering alarm meaning that it needs to detect the object consistently as a person for say more than 3 or 8 seconds. A valid human will be seen as a human from all angles. A deer only from very select angles. Upping the sample time reduces the chances of a deer triggering a false alarm. Unlike some other vendors, I am *never* going to claim that you can ever get to 100% accuracy in all cases because that is just simply impossible. Hell, it's hard enough to claim 100% reliability from a simple burglar alarm door contact or motion sensor. To claim it for any advanced sensor system like this is just dishonest, IMO. My personal opinion is that our system will provide the best possible results for visible or thermal cameras. BTW, the car wash video clip is from one of the VideoIQ cameras at my house, and came straight off the system, there was no trickery or editing there. Here is another clip of me and one of the dogs at the office: The yellow box you see briefly on the dog is what we call a "suspicious" or "unclassified" object. The system sees that the dog is starting to match some criteria, but not enough for a full classification.

Less is more, meaning that one switch would be the preferred method in this case.

I get foxes across my property fairly regularly. Except that *my* analytics generally ignore them Animals that are not uncommon to see on my property here (southern NH): Foxes Turkey Moose Coyotes (bastards) Black Bear Porcupines Last year we had a fox that came by almost every morning around 10:30, I'd see him trot across the driveway frequently. The moose are mostly late fall and early spring. Turkeys are usually in the fall.

You'll need to have the client-side router establish a VPN. Either to a VPN service, or to a VPN terminator that you own/provide some place. You're basically behind a giant ISP firewall that you have no control over.

Hi... I run the Field Engineering group for VideoIQ worldwide. My job is basically to make sure things work in real life and that the sales people stay honest First off, you're not going to get anywhere near the quality/reliability of this solution with PIR's, DVR/NVR software, "advanced motion detection" or any of the other relatively unsophisticated "solutions" on the market. We're also the *only* ones doing real megapixel analytics, meaning truly analyzing the full stream, not downsampling it and running an analysis on sub-sampled images. Your application fits the description of probably 80% of our typical deployments. Not the bit about the farm, but the bit about having a tricky location where you need something that can truly help deter crime. I've got a really good distributor based in the UK (where it seems like you are). PM me your info and I'll put you in touch with them. I'll also give you my corporate contact info and you're welcome to contact me as well with any questions or whatever.

Thanks for the reply. This is helpful. This is just a matter of personal curiosity about a criminal case which was covered on TV. I've been reading a book about the case. The video from the camera in question figured heavily in the conviction of this person. The time on this camera was 12 minutes behind the hotel system CCTV camera time. It apparently was rigged to spy on this particular person. It was right outside his room and focused on the door. I assume that it was monitored from someone's computer as it was not a part of the hotel's regular security system. I'm aware of several hotel scenarios where cameras installed for a specific purpose (eg: to monitor a specific guest or area) are NOT part of the standard CCTV system. These cameras are intended to be covert in appearance (though most are easily spotted hacks like this one) to both guests AND the standard security personnel. There will be a very small group of people aware of the camera, depending on what is being investigated. If it's an internal hotel issue (eg: suspected housekeeper theft) then only the head of security may be aware of it. Recording could be done within the camera itself (SD card), to a small mini DVR or NVR device installed in the same enclosure as the camera, to a DVR/NVR that is near the camera (tossed up in the drop ceiling is common), or to a remote recording device carried as an analog or IP signal. In several of the cases I've been involved with if an IP camera is used it is kept completely off any normal networks to prevent it from being accidentally discovered or tampered with. Given the vast supply of products on the market it would be silly to speculate the state of components in a generic enclosure. This is the Schroedingers Cat of security camera systems It could be an analog camera, an IP camera. It could be just an image sensor cabled back to the "camera" electronics (seen that done more than once), it could even be a small FLIR TAU module (unlikely in this case, but possible). Since you don't want to supply any more details of the case, the book, the TV coverage, etc. the only thing anyone can say is that it could be anything you can possibly imagine.

We'll be there exhibiting.

A few responses/thoughts. POE - Just FYI, POE has the same basic limitations as Ethernet (100 meters). If PoE won't work on your cable run, you're probably exceeding the Ethernet distance limitations as well. Resolution for court of law - There is no standard for this. Video evidence is usually part of a court case, but in reality the goal should be to prevent stuff from going to court in the first place (eg: deter theft, not just record it). Resolution to recognize faces - This will be somewhere around 80 pixels per foot during the day, and 100 pixels per foot at night (you will typically need more resolution at night because the lighting is reduced, and so then the contrast/details are also). So, you can take a camera image sensor and divide the horizontal pixel count by 100, and that is how many feet wide that image sensor can cover and give you really good detail. I think you will also find that it's going to be cost prohibitive to get that much detail everywhere, so you'll end up with cameras at choke points (entrances/doorways) that capture strong details, and the other cameras will basically be overview cameras. Cameras with motion detection - This generally won't work the way you want, it will generate too many false alarms. You can accuse me of bias here since I work for a company that specializes in video analytics, but our analytics are generally well regarded... Limited lighting at night - Cameras see light, there is no way around this. No light = no image. If you don't want to put up visible lights you can consider using IR illuminators. Beware of cameras that offer too-good-to-be-true low-light specs. This is typically achieved by binning (reducing the resolution to ~1/4 of the daytime resolution) and/or digital slow shutter (taking a longer exposure, which results in objects moving (like a person walking) being blurred to the point of being unrecognizable). D1 cameras (IP or Analog) will generally have better low light performance than megapixel cameras, but of course much lower resolution as well. You might want to consider doing this project in phases. There are a lot of variables here, but to really deliver what you want, a $5,000 budget probably won't cut it. If you have a high level of threat, you might want to consider using a video central station that can receive alerts after hours and act on them. If it's low threat, you might be able to self-monitor with email alerts (though this can be an admitted pain in the ass...). You *could* consider having some additional network drops installed that would allow you to relocate cameras based on your current inventory in the yard so that way if a large vessel is blocking a shot you could move that camera. It's easy to get an IP66 rated camera (can handle outdoor environments, rain, splashes, etc.), but somewhat more difficult to get a good IP67 camera that can survive a direct spray from a wash-down hose, but there are lots of factors here. I would just count on making sure your cameras don't get hosed down directly.

You'll *probably* find that you are better served with a camera that has an alarm or contact-closure input, along with the features you described. There are actually many options. Then, select an appropriate PIR sensor, and place it accordingly. In many cases the optimal camera and PIR locations are not the same, so you end up compromising one or the other. Keep in mind too that for outdoor scenarios this approach can be highly prone to false alarms, especially as you try to aim for longer/larger detection areas.

I understand that you may not be in a position to turn down certain work, but I think you will find over time that the "budget" jobs that are extremely cost-conscious are often the least rewarding overall. Spending hours trying to figure out someone else's wiring and and work can be a frustrating and time-consuming process.

Have some vinyl stickers made up that say "Verizon" "Comcast" "High Voltage", etc. Then, just affix a sticker to a random NEMA box. Some things for this to look believable... Use a decent quality sticker, but don't make it TOO fancy. Don't use anything you can buy at Home Depot for the enclosure, order something online that looks rugged or just uncommon. Verizon wouldn't go to Home Depot and buy a $15 NEMA-4 plastic junction box to hold their equipment... Store these outdoors when not in use, or otherwise find a way to make them not look "new" when first installed so that they don't catch too much attention. The problem is that what you are asking for is a relatively low-volume business. So, yes, you could build something cheaper yourself, because when you buy it the original seller needs to make enough margin on it to make HIS efforts worthwhile. It's kind of a chicken and egg problem. There is not to my knowledge any reliable source of "empty covert housings to put cameras in". Everybody just kind of improvises.

I've seen similar cable bundles used for a decade at least. Usually something like 2 or 3 RG-6, 2 Cat5's, and a 18/2. Some of the newer ones also have a pair of fiber strands. Belden is one company that makes such a cable, it's usually listed in their residential structured wiring products list.

Buy a basic DIN rail mounting plate for it?

100VA units are going to be a little tough to find in a DIN rail mount, though Digikey has them. Most of the units I've seen have been 3A at best, and most are around 1A total (eg: the ComNet one). So, you might need to use multiples.

Thats kind of what I thought was giving me trouble, the nice part is though, having wireless around the shop too. This 4o4 8 port switch would be like a router? split the signal from the ubiquiti? I have the secrurity system, plus a computer out there that i am running off the ubiquiti. I am just using the ubiquiti's in bridge mode. Sorry if I am frustrating you guys , I would love to have it work with the hardware I have right now if thats possible? Sounds like you got it worked out. Just wanted to clarify, I meant to write "4 or 8 port switch", not 4o4 8 Port switch. Also, just from a simple "training" perspective, you don't "split" an Ethernet signal. Ethernet is designed to be a shared network for access by multiple devices. A switch is the "splitter", the reason you need a switch and not just a simple device like a phone splitter is because there are timing concerns and other things that happen to make sure two devices don't try to talk at the same time and so forth. If you have an Ethernet cable carrying a signal from a device like a router, wireless bridge, or another switch that is going into a computer, you can plug that cable into a 4 port switch (as an example) and then get 3 new ports to plug more devices into (1 port for the uplink from the existing cable, and then 3 ports for new devices). There are certainly a lot of other factors to consider, but for a simple home network you don't have to get too deep into design theory and advanced considerations.

Most expedient route might be to toss the Dlink and go buy yourself a simple/cheap 4o4 8 port switch. Without getting too far into the details, you basically have 2 separate networks, the one at the house, created by the router there, and the one in the shop created by the dlink. For this to work easily, you want 1 network. The ubiquitis are like an "invisible ethernet cable". If you were doing this *without* the wireless part, you'd just put a switch in the shop, not another router. And that is going to be the easiest solution, but not the cheapest.

It *is* more expensive, I'll agree with that. An RJ-45 plug is a few pennies, a decent keystone outlet is at least a couple of bucks. I think your experience is not indicative of the overall industry though. My personal experience, in dealing with cable plants since before Cat 5 was a standard, is that field-crimped RJ-45s are one of the leading causes of network problems, and punched-down cables overall pass more certification tests and offer a longer-lasting solution. In both cases though, I'd say that the plural of anecdote is not "data". Organizations like BICSI that are dedicated to these issues have recommended punched-down terminations on both ends of your solid-core backbone cable for what are, IMO, good reasons. You're not arguing against *me* on this, you're arguing against statistics gathered over millions of cable plant installations.

Not sure who is "getting bent out of shape"... Yes, the RJ-45 connectors aren't exactly industrial-grade to begin with. So let's tempt fate even more by using a non-recommended termination method. You're acting like punching down a simple keystone jack is rocket surgery or something. I'd be willing to bet that if you polled most full-time professional cable plant installers and asked them to spec a job with crimped-on RJ-45's, they'd either refuse, or would add a clause that nullfies, or significantly reduces, their warranty and claims around certifications.

Of course not. But wouldn't you agree it should at least mean "done 95% of the time", or some similarly large majority of the time? Why would you *not* want to follow basic best practices?

Will it still work in 5 years? Is it serviceable? Will it still work if the connection gets flexed or moved significantly? If the customer has problems with the install that are not necessarily related to your work, will someone be able to come along and call the cable plant into question because the terminations do not conform to industry best practices and specifications? Your binary view of "works" or "doesn't work" does not, IMO, take the larger factors into consideration. While I don't think you're going to be doing any enterprise-class jobs right now, I still think it is worthwhile to do the jobs properly. Solid-core cable gets punched-down, not crimped with RJ-45 ends.

How so? It's important to know how to do a decent RJ-45 termination for the time when there is no other option. But if you're doing them regularly, you're doing it wrong.

Agreed. It's good to be safety-oriented, but I don't think there is any free-standing ladder that is going to reach warehouse ceilings that I'd trust *and* want to carry around a jobsite all day. If they *don't* have a lift, then rent one, and it goes into the job cost. If they don't have a lift and the job won't permit the ~$1200/day cost (at least here), this might be a good job to refer to your least-liked competitor...

I wouldn't spend that much time and effort practicing something that you probably shouldn't be doing in the first place...