Russ: This is the BusinessMakers Show heard here and seen online at theBusinessMakers.com. It's guest time on the show and I'm very pleased to have with me this morning Ray Davis, founder and CEO of Behavioral Recognition Systems. Ray, welcome The BusinessMakers Show.
Ray: Thank you, Russ. Thanks for having me.
Russ: You bet. Well let's start by you telling us about Behavioral Recognition Systems.
Ray: Well Behavioral Recognition Systems, the short BRS Labs, invented the only way to connect video cameras to a system that's widely called machine learning. The technical term would be artificial neural network which emulates the human brain and the way it thinks and applies reason to what it's dealing with. Until now, no one was able to connect a camera to this type of technology so they created a rules-based technology in the surveillance industry which allowed them to decide what types of objects or behavior they were looking for. In short, we turned around and took that a step further and connected it to this artificial neural network and allows the computer to watch the cameras and allows it to decide what is normal and abnormal. And in doing that, we're able to raise the situational awareness and to give more valid alerts to the security staff in the field and both in their center.
Russ: Okay, well I have a little bit of experience with neural networking and so you're talking about learning software.
Ray: That's right.
Russ: That watches behavior and then starts pointing out what would be abnormal as opposed to prior to BRS, companies would sort of highlight - these are things that we think are abnormal behavior that we would like to monitor with video cameras.
Ray: That's right and the main problem with that technology is, of course, it's limited in scope in that, you know, if you don't select or decide what it is you're looking for, you're not gonna catch it.
Russ: That's right.
Ray: Number two, because it's so specific, it's very prone to come up with false alarms.
Russ: Okay.
Ray: If you draw a line and say if someone steps over this line, give me an alert.
Russ: Right.
Ray: Very often you'll get, you know, shadows. When you're working out in the real world, you have shadows -
Russ: Right.
Ray: - glares, all these things that set off these false alarms.
Russ: Right.
Ray: And then, of course, one of the major problems with it, too, is the installation. It can take from five hours to five days per camera to set up a system. So consequently, very few video analytic systems are out there of any size. I think one of the largest ones is 60 cameras.
Russ: Okay.
Ray: When with our system, we can install within two hours on whether it's 10 cameras or 10,000 cameras.
Russ: Okay. Now but with your system, with the BRS System, I would assume that it takes a period of time for the system to go through its learning process and know what might be abnormal behavior.
Ray: That's correct. I mean the longer it learns the better it performs. Now it does have a point where it becomes self-aware. So it's typically about two weeks. It can be as few as five hours, but what we tell our clients is to let it run for two weeks. Each camera will decide when it has had enough knowledge base to make valid alerts and at that point, it'll start giving you those alarms. But in all, you're really looking about a two-week training period.
Russ: Okay, talking with Ray Davis, the founder and CEO of BRS Labs. Now Ray, I keep talking about behavior that it's monitoring and in my mind that means it's just monitoring and watching human beings. Is that the primary and the total extent of the software's ability?
Ray: No, the software actually watches every object in the camera's view. It understands the behavior of a chair, of a briefcase. It classifies objects into thousands of different categories, both recognizing their size, color, shape, even weight, and understands how that behaves. So instead of having to tell something to watch for an object left behind, it can actually recognize that occurrence even when no one has told it to look for it because it is able to watch, not only the human carrying the object but the object itself and understands how it behaves.
Russ: Okay. Well, I just imagine with the security issues of society here in 2010 and the whole terrorist thing, that it could be an extremely popular product and service that you offer today. Would that be accurate?
Ray: It's very accurate. I mean, we made it commercially available September of 2009 -
Russ: Okay.
Ray: - after over 150 man hours of development.
Russ: Okay.
Ray: Four years.
Russ: Okay.
Ray: But since then we have installed over 30 installations all over the world and in every one of those, the system is performing very well. Everyone that has it has stepped up and has expanded their system; has given us great reviews and loves the system.
Russ: Okay. Now when you first explained BRS Labs, you said you're one of the first software applications that connects video imagery with neural networking. Are there others now, competitors in this space that are close by, yeah?
Ray: There are competitors in this space; probably two dozen.
Russ: Right.
Ray: But they all use what we call the rules-based technology.
Russ: Okay.
Ray: Ours is the only one that uses this reason-based technology and we have filed the patents on that so we have over 40 patents on that technology now. So you know, we pretty much hold that technology to ourselves for the next 18 years.
Russ: Okay and once again, how many installations did you say that you have -
Ray: Over 30.
Russ: Okay.
Ray: And these are mostly government but some commercial applications.
Russ: Okay. Now, I'm just curious. Once you turn over the product to your customer and they're using it to monitor behavior, are you still connected into the behavior that they're monitoring or is that totally independent of BRS Labs?
Ray: It can be any way you want it. I mean, the system is designed to where the user can install it in, you know, in Brazil. For instance, we have it installed up to 20,000 different ATM machines.
Russ: Wow.
Ray: And in some situations you don't even have an Internet connection. So it's a totally local system whereas here in Houston we have a company called Ceva Transportation - third largest transportation company in the country.
Russ: Right.
Ray: They have over 5,000 cameras locally and they run it all centrally to their location and monitor the cameras from all over the country.
Russ: Very interesting. Very interesting, cool. Well I'm talking with Ray Davis, founder and CEO of BRS Labs and I'll be back with more with Ray after this. You're listening to The BusinessMakers Show, heard here and seen online at theBusinessMakers.com.
[Aflac Commercial]
Russ: This is the BusinessMakers Show heard here and seen online at theBusinessMakers.com and continuing on with Ray Davis, founder and CEO of BRS Labs. Well obviously it's a happening area right now but I'm curious how long have you been doing this, Ray?
Ray: Russ, we've been in business since November of 2005.
Russ: Okay.
Ray: It took us the four years to actually develop and perfect the technology and then we made it commercially available in September of 2009.
Russ: Okay, so you went four years without customers, without revenue.
Ray: That's right and it is pretty much against, you know, Business 101 because what you want to do is get out there and grab market share and start covering your burn as fast as possible.
Russ: Right.
Ray: But that's pretty much what our competitors all did. We wanted to sit back and actually create the technology the right way and make sure that when we did launch the system, it was applicable to all of the different verticals that we approached.
Russ: Okay. Well tell us what you can tell us about that first customer. How did you capture 'em and how did they trust you and that your system would work and how has that gone?
Ray: The first true customer was the Marriott Hotel in Mumbai. If you remember a couple of years ago -
Russ: Oh yeah.
Ray: - they had the terrorist attacks there.
Russ: Right.
Ray: They found us through the U.S. State Department. The Department of State actually brought us to meet with them and within ten days we were actually up and running in that hotel and were able to start giving them valid information and thankfully, we haven't had any problems there since.
Russ: And that was after the terrorist attack?
Ray: That's correct.
Russ: Okay and so wow, interesting. Okay.
Ray: Yeah, that precedes our September launch by a bit but they were pretty adamant about needing the technology as fast as possible.
Russ: Okay. Now would - by chance would you do any sort of observation with your software in airports?
Ray: Airports? Yes we have the Dubai Airport, for instance.
Russ: Okay.
Ray: They are very concerned about not only terrorists getting into their baggage systems but also children becoming trapped in those systems.
Russ: Wow.
Ray: So we're able to watch the belts there from the minute you hand it off to the agent to the minute it gets on the plane and we can determine exactly what and where those bags are and how the objects behave and tell the difference between a suitcase and a child or a person. We even tell the difference between the staff because in that airport the porters tend to ride the belts themselves. It's a long beltway there. So we have to tell the difference between a uniformed employee and non-uniformed or a child.
Russ: Okay. I'm curious about the level of sensitivity. I mean can your system actually notice abnormal behavior in an individual? Or do they just identify individuals that might need to be watched and monitored?
Ray: Well we have a system called a microclassifier which we also patented which, you know, looks at the color, the hue, the shininess, all the relative points about an object and then we're able to classify it not only is it a human or what type of object it is -
Russ: Right.
Ray: You know, there are over a dozen patents out there on how a human can be identified on camera but you know, that's too specific. You have two arms, two legs, but if you have a camera mounted 30 feet in the air -
Russ: Right.
Ray: -it's hardly even able to see the legs.
Russ: Right.
Ray: Each camera has to learn individually what each object looks like in that camera's view. And that's how we're able to actually classify these objects so precisely.
Russ: Okay. Now I understand you have installations internationally, correct?
Ray: That's correct.
Russ: And now I also understand that in the U.K., now particularly in London, they use cameras extensively to monitor everything that's happening on the streets. Do you have any installations that are similar to that?
Ray: Yeah, in the U.K. you have, you know, millions of cameras -
Russ: Right.
Ray: - just in London alone and, you know, we launched our London office at the IFSEC Convention there in London and we have quite a bit of business in London. That is a hot spot for video surveillance.
Russ: Right.
Ray: I don't wanna say hot spot for terrorism but they are very aware of their surveillance system and the fact that just using surveillance cameras alone isn't enough because they've spent billions on this surveillance system but their crime rate -
Russ: Right.
Ray: - actually has gone up.
Russ: Right.
Ray: So with a system like ours it's able to proactively catch behavior and in many cases catch a terrorist or a criminal event even before it happens.
Russ: Oh wow. In fact, I'm curious about this. You mentioned while ago that it's difficult if you have a camera that's mounted 30 feet in the air and monitoring a whole group to really capture some of the detail but perhaps is the eventually gonna evolve to the point where there's cameras down low and multiple cameras everywhere watching every move?
Ray: Well, you know, in London you do have cameras everywhere at every level. Here in the 'States, I think in many installations you'll find cameras primarily installed at the top of the buildings which is gonna give you a very low quality as far as looking down at subjects.
Russ: Right.
Ray: You're not gonna be able to recognize them.
Russ: Right.
Ray: But more and more we're seeing businesses and government becoming more aware of the angle, the view of the cameras and making sure they get that face, you know -
Russ: Right.
Ray: - when something happens.
Russ: Do you have any push back from those that are so sensitive to privacy these days?
Ray: No, we actually did a study a year and a half ago and found out that, you know, 96, 97 percent of the people have no problem with that. The problems occur when you're able to identify the person. Now we identify the object as human, automobile, train, truck, car, whatever, but we don't give you the name of the individual. So we're tracking the behavior of a human, not Ray Davis, you know, walking down this specific street at this specific time.
Russ: Okay. Talking with Ray Davis, founder and CEO of Behavioral Recognition Systems and we'll be back with more with Ray after this. You're listening to The BusinessMakers Show, heard here and seen online at theBusinessMakers.com.
[Aflac Commercial]
Russ: This is the BusinessMakers Show heard here and seen online at theBusinessMakers.com and I'm talking with Ray Davis, founder and CEO of Behavioral Recognition System and Ray, before we went to the break, you were talking about not offending those people worried about privacy because you don't really ever identify the individual. Well let's say that there is an individual, a terrorist, that everybody knows and knows generally what he or she looks like. Wouldn't it be fairly easy to, in your program, identify the aspects of that person's image and look for them?
Ray: It would be. That's, let's say an attachment to our system that we haven't put in because of the issues with privacy, but there are significant companies out there with facial recognition products that would attach an interface with our system that would give that solution to the user.
Russ: Okay. Well I find your company and this whole area quite fascinating. What do you think the future's gonna look like in this area of identifying behavior through recognition systems?
Ray: Well I think eventually the behavioral recognition will become just as prominent as voice recognition. I mean, as it is today, you talk to your car. You call your bank and a computer talks to you. I think it will be in every aspect of our lives and will become commoditized over time. You know, in 20 years, people are not even gonna be aware of the situation. It's just gonna be a given that that technology exists and we're being watched by computers.
Russ: Right, right. That's really interesting. Well before I let you go, wanna know a little bit more about you and your background. From what I understand, Behavioral Recognition Systems is not your first rodeo, right?
Ray: No, I'm - I am what you'd call a serial entrepreneur.
Russ: Okay.
Ray: This is my eighth company.
Russ: Eight, wow.
Ray: In 35 years. I typically come up with an idea, fund it initially, find out if it's commercially viable and then bring in investors and build out the product and then have an event where you would sell the company or go public.
Russ: Right. Well in fact, in that regard, what's the status of Behavioral Recognition Systems? Do you have outside investors?
Ray: We've raised a little over $52 million to date.
Russ: Okay.
Ray: We have outside, primarily high net worth individual investors. You know, our resellers are the companies like Cisco and IBM and Unisys, all types of companies that would be able to acquire this company for its technology. That is in the cards, probably, for a larger company with their own network and their own sales staff that would be able to take and take advantage of this technology.
Russ: Okay.
Ray: And take it to the market themselves.
Russ: Okay. Well I really appreciate you sharing your story with us.
Ray: Thank you. Thank you for the time. I appreciate it.
Russ: You bet. We've been talking with Ray Davis, founder and CEO of Behavioral Recognition Systems and you're listening to The BusinessMakers Show, heard here and seen online at theBusinessMakers.com.
