Video Analytics Implementation Challenges + Opportunities

In our blog, we have demonstrably argued for some benefits and application of the Video Analytics technology. For successful application of Video Analytics however, empirical evidence shows that several critical factors that need to be taken into account.  According to Aimetis, a video security and surveillance management company, lists environmental factors and computational factors as the key determinants of a successful use of Video Analytics. Among others, camera angel, weather conditions, background lightning levels can all impact the video analytics algorithms. In addition, computational factors such as processing power of the recorded segments, resolution and frame rate as well as storage capacity all play a vita role in the final delivered value to clients using a Video Analytics service. Nonetheless Aimetis also states that: ”repeat occurrences of false alarms, combined with undetected events, can significantly deter end users from using video analytics”. 

With regards to previous application of Video Analytics, the absence of sophisticated processing hardware, autonomous and intelligent camera that can track movement transformed today’s application of Video Analytics to a more conventional and labor intensive practice. For example, in a secure environment using cameras, previous practice would require a classic situation room, with security personnel to monitor and judge the video content. While Video Analytics does rationalize some of the costs, the benefits of Video Analytics is still in continuous development. Environmental factors described above are difficult to control while cost of processing and detecting desired result from video analytics could equally depend on the funds a company is ready to invest in the technology and devise how such cost can offset conventional video surveillance methods described above.

However, as the world continues to move toward an ever more sophisticated information society, the development and forceful implementation of Video Analytics Technology is here to stay and keep evolving toward higher value and delivered results for its clients.

How far can we go? (Video analytics vs. privacy week)

The application video analytics is perhaps most often associated with is security. For some, security is the ultimate goal, the only true value. But for others, improvements in security have to be weighed against other values, such as privacy. As video analytics becomes everyday and common, societies will eventually have to take a stand towards how far can we go? When does automatic video surveillance, face recognition etc. become intrusive? Should the surveillance data be stored? Who has access to it? What should it be used for? Many questions remain unanswered, and it’s hardly surprising that security companies and civil rights groups are at odds here. In any case, the discussion that will take place during the next few years will be very interesting and will shape the society of the future. Technology will eventually enable absolute connectivity that will make our lives easier and mostly better. On the other hand, this can result in almost total loss of privacy. Finding a balance is likely to prove to be a tricky task.

How about privacy?

Our theme of the week: Video analytics and individual privacy

Here are two critical views on the topic, more to follow.

Whats wrong with public video surveillance?

Big brother – you may be watching but how much should you see?

Let The Boston Tragedy Be The Last

The Boston Marathon Bombing left a scar in the hearts and minds of the people and reminded us that we live in a world where death might be just around the corner.

As already mentioned in the earlier posts, Video analaytics is a tool that is widely used for security purposes. In the case of the recent bombings, Video analytics was used in the investigation process to identify the suspects and locate the origin of the blasts.

The main point of this post is not to praise video analytics that helped capture the suspects in a manner of days.

With the use of video analytics we have a chance at preventing such event from happening.

Experts suggests that VA can be used to identify unattended belonging as was the case in Boston.

Let’s hope that the speed of VA development is faster than the speed of potential terrorists using new technology to unleash a reign of terror again.

How Intelligent Video Analytics (IVA) is changing video surveillance

Video Surveillance has been used as a security tool for a long time. The idea of having a control room from where guards or operators could watch what was happening at different places at the same time made patrolling activities cheaper and more comprehensive. With time, more and more cameras (and monitors) started being used in order to ensure that every inch of a building or an area was being covered by the system. However, the vast amount of images generated made it complex to manage all that information. As it became increasingly harder for humans to deal with such high quantities of different images in different screens at the same time, machine intelligence then was introduced to facilitate data management. The newest thing today is the use of Intelligent Video Analytics (IVA) to help in real time event detection and also in post event analysis. These might include motion and intrusion detection, asset protection, pinpoint an event in recorded video, etc. This technology helps to increase productivity and efficiency of video surveillance systems and the people who monitor them, focusing on automating video analysis and security alerts, therefore reducing the need for manual work and also decreasing monitoring costs.

IVA is becoming a vital component in the growing surveillance systems industry and it is projected to be the most important technology segment of that sector. The IVA market is expected to grow to $867.8 million by 2017, according to some studies. 

Check out the videos below to see a IVA system working.

Shopping for diapers or condoms?

A strong litmus test for shoppers: Do you enjoy the small walk from one escalator to another while shopping in a store? If yes, you are the ‘I don’t wanna miss anything’ kind of shopper. If no, you are the ‘I need to finish my shopping list’ kind of shopper. Shopper you are, in any case.

However, identifying these little subtleties in consumer behavior can be of great value to retailers. And that’s why stores are extensively using video analytics to unlock information on customer patterns, trends and behaviors. 

Let’s say a store positions diapers next to condoms. Through video analytics, retailers can identify consumer behavior in terms of how long customers take to decide before making their purchase, whether they make the purchase with/without kids, the speed with which they walk through each section, their gender and their age.

Not only that, based on the volume of traffic entering the store, retailers can be alerted of the need to open additional tills proactively before queues start to form. Also, by counting the number of people in each part of the store, video analytics can locate ‘hot zones’ – those with the most customer activity – to help maximise in-store promotional campaigns, and identify ‘cold zones’ to determine how store layout changes can improve customer traffic flow.

So next time you visit your local store that suddenly has positioned diet products next to swim wear, there is a good chance that they know you are trying to lose weight for your vacation to Ibiza this summer!

I was Busted by Video Analytics

Recently I went on a roadtrip with some friends to south Spain.  We rented a car for a few days, went down to Sierra Nevada for some nice skiing, and returned the car in Madrid happy that in over 700 Kms. driven we hadnt gotten pulled over by police. Talk about celebrating too early....

SURPRISE SURPRISE, 3 weeks later a 100 euro speeding ticket arrived in my mailbox. 

Those damn traffic cameras (see one below)! 

I had forgotten that with Video Analytcs technology you dont need to get pulled over by a police car to get a ticket!!!

Video Analytics software is allowing municipal governments around the world handout tickets to  speeding drivers without any police intervention.  The software continuously scans the video being shot by the cameras when it detects a vehicle is passing, it uses an algorithim to determine the vehicle's speed, and if  they are speeding, it will snap a shot of the vehicle's license plate and trasmit the data for the ticket to be issued.  All in all the ticket arrived in less than 2 weeks after the incident!  As you can see the camera got a nice shot of my rent a car and included the details of my speed, time and place where i got caught, pretty impressive! 

 

So i learned my lesson the hardway, video analytics does work and now i know to stay at or below the speed limit even if there are no police in sight! 

In addition to being used for detecting speeding drivers, video analytics is also used to detect vehicles going the wrong way, drivers that burn red lights or are parked incorrectly and to provide statistics on traffic patterns. 

To see more on how video analytics is assisting with Traffic Management please follow this link:

http://www.traficon.com/page.jsp?ref=principles&id=8&parentId=3

Smart way to save on car insurance deductible costs using VA

Adding on to the commercial use of video analytics in the automotive sector, Audi, the German automaker, recently launched a 360-degree Camera Assist Parking System. Previously launched by the competitor BMW, the technology enables the driver to follow the vehicle’s exterior movements through the dashboard video screen which projects a live top-down view of the vehicle. While the vehicle itself is projected as a ”dummy” image, the vicinity footage of the vehicle is fused together by four wide-angle cameras which are embedded in the front, the back and to the sides of the vehicle.

Although other safe-parking features—acoustic parking sensors and rear-camera parking— have been in the personal vehicles market for some time, the 360-degree Camera Assist Parking System is a revolutionary video analytics technology. The system can help reduce the risk of body vehicle damage during parking or when driving through narrow city terrains, which ultimately can reduce unpleasant insurance deductible costs.

The camera assistance is optional for for the 2013+ A8 and S8 models which belong to Audi’s largest sedan vehicle range with the least cockpit visibility. 

Lights, Cameras, Revolution

“The NBA is undergoing an analytical transformation, and the Raptors are one of the teams at the forefront. For the first time, here is an exclusive, inside look at how SportVU is changing basketball.”

The most important innovation in the NBA in recent years is a camera-tracking system, known as SportVU, that records every movement on the floor and spits it back at its front-office keepers as a byzantine series of geometric coordinates. The Raptors' analytics team wrote insanely complex code that turned all those X-Y coordinates from every second of every recorded game into playable video files. The code can recognize everything; when a pick-and-roll occurred, where it occurred, whether the pick actually hit a defender, and the position of all 10 players on the floor as the play unfolded. The team also factored in the individual skill set of every NBA player, so the program understands that Chris Paul is much more dangerous from midrange than Rajon Rondo, and that Roy Hibbert is taller than Al Horford.

The ability to recognize individual player skills is crucial for the juiciest bit of what the Raptors have accomplished: The video demonstrate some “ghost players”, and they are doing what Toronto coaching staff and analytics team believes the players should have done on this play. The system has factored in Toronto's actual scheme and the expected point value of every possession as play evolves.

Violence, Parenting, and Video Analytics

Raising a child is a painstaking task.

Not only is the parents responsibility for leading the child to a bright future, they are in charge of shielding negative forces that may have a harmful impact. I would like to focus on one such forces, violence, and how video analytics may help protect children from growing up thinking violence is a part of daily life. Here I am not referring to physically guarding children from violence. Instead, video analytics can help children from being the perpetrator of violence.

As some research suggests, exposure to violent TV programs and movies makes children behave more aggressive and increases the chance of them being convicted of a violent crime in the future. What video analytics can do is to detect and remove violent scenes by identifying blood pigments as suggested in an article titled “VIOLENT SCENE DETECTION IN MOVIES” published in International Journal of Pattern Recognition and Artificial Intelligence. By being able to filter the content, children are saved from exposure to violence even when they are out of the reach from their parents. 

Thanks to video analytics parents have one less item on their parenting checklist.

Violent scene from a movie    

(Source: “VIOLENT SCENE DETECTION IN MOVIES”)

Video analytics identifying blood pigments from the same scene

Revolution of movement (powered by video analytics)

Nintendo Wii’s launch in November 2006 changed console gaming for good: The controller, “Wii Remote”, could detect user’s movement. For the first time, gamers could interact with the machine via a materially more natural interface: their own movement. The user interface of Wii, however, still included a controller. Something the gamer had to hold in his/her hand. And the motion detected was only that of the “Wii remote”. Even though the leap from a traditional controller, mainly operated using gamer’s thumbs, was gigantic, the interface still remained one step away of purely free movement and fully natural control.

Video analytics changed this. 4 years later, in November 2010, Microsoft’s Xbox 360 Kinect was launched. Instead of requiring the user to hold a controller, the interface was purely based on spoken commands and gestures – not only those of a single hand, but of every part of the user’s body. Utilizing video analytics, Kinetict’s sensor device was able to accurately detect gamer’s movement and to turn that into a completely new gaming experience.

Wikipedia’s list of already launched and upcoming Kinect games includes titles with themes ranging from golf to Angry Birds and from football to Harry Potter. The number of possible applications is almost infinite and the future of gaming looks more interesting than ever as more and more game developers start exploring the new possibilities. In future posts, we’ll be digging deeper into what to expect from the revolution of movement in gaming and will have a look at how video analytics and gaming consoles can be used for medical purposes. Stay tuned!