# Appropriate Applications for Megapixel Cameras

## Although megapixel cameras may not be necessary for all applications, there are many where their technology provides an undeniably superior solution. Features like greater detail and wider coverage make them ideal for identification and forensic purposes.

Many applications require positive identification of vehicles or license plates in parking areas. To determine coverage or the number of cameras required for any application we can utilize the following math. First, we need to know the size of the area. Let’s assume a dimension of 200 feet wide for the parking area. In order to identify the license plate, we need approximately 40 to 50 pixels per foot.

We then perform some calculations to see that 200 feet X 50 pixels = 100,000.

Thus, using a 3-megapixel camera with a horizontal pixel count of 2,048, we would require ~5 cameras (10,000 / 2,048 = 4.9) to properly cover the area with a scene width of ~ 40 feet.

With this information we can now select the lens for our application. Those who have worked with CCTV should recognize the following formula:

Lens = distance / subject
X sensor format

So the lens selection should be based on the distance the megapixel camera is located from the scene. Remember the scene width cannot exceed 40 feet.

Referring to the “Lens Choice for Identification” chart, at a selected distance of 100 feet an 11mm megapixel lens would be required when using a megapixel camera with a sensor size of 1⁄3-inch and a 16mm megapixel lens for 1⁄2-inch sensors.

Sensor sizes or formats for megapixel cameras range from 1⁄3-inch to 1 inch (in the photographic world) with 1⁄3- and 1⁄2-inch formats being the most common in the CCTV industry at the present time.

Application 2: Airports
Increased efforts to protect critical airport infrastructure following 9/11 have led to increased applications to combat security threats and protect airport property. However, often budget cuts prevent the necessary system infrastructure from being put in place and many times the number of cameras required for the proper surveillance to protect these areas also falls short.

To help reduce costs, megapixel camera systems can use existing infrastructure. For locations that have an existing local area network (LAN) the installation of an IP megapixel security camera system might even work out to be less expensive than a conventional surveillance alternative. Instead of running coaxial and power cables across long distances, only very short runs of affordable Cat-5e or -6 to the closest network data points are necessary.

Even if the installation of a LAN should be needed at your project’s location, the installation is much less expensive compared to massive coax runs. When
designing system applications, make sure you follow the rules and bandwidth requirements to accommodate a megapixel camera. (See the next issue of “D.U.M.I.E.S.” in October for more information on megapixel connectivity.)

Application 3: Video Analytics
Integrating today’s megapixel cameras with video analytics can improve the overall strength of the surveillance system. Traditional security systems are not always able to identify events in real-time. Standard CCTV systems provide only passive recording capabilities with minimal active monitoring. [IMAGE]dumies8-3.jpg[/IMAGE]

As a result, these systems are used for after the fact, not prevention. Not using video surveillance to its full potential as a proactive or realtime detection system is unfortunate.

Video analytics with megapixel cameras is an excellent tool in the fight to protect critical applications. No matter how highly trained or how dedicated a human observer, it is impossible to provide full attention to more than one or two things at a time; and even then, only for a few minutes at a time. A vast majority of surveillance video is permanently lost without any useful intelligence being gained from it. Many system operators have missed critical information due to fatigue and plain old boredom.

The video surveillance industry is currently challenged with a growing need to assist surveillance camera system operators and designers in improving image quality and adding intelligence while keeping overall system costs down.

So what can be done today? License plate recognition has been around for a few years and is accepted; however, it is not always 100-percent accurate. The introduction of megapixel technology has greatly improved the accuracy of license plate readers as well as provided a means by which many of today’s systems are completely automatic.

Megapixel cameras have also revived the facial recognition marketplace. Face recognition is notoriously difficult to perform reliably, and is extremely easy to fool by using disguises. Plus to work with any degree of accuracy, an excellent headshot of the subject is required. High-resolution sensors will open up new possibilities on these fronts.

In addition to the familiar advantage of depicting more detail with greater pixel density, a less obvious development stems from the inherent flexibility of digital image formats. Not only do new sensors support the 16:9 display perspective and similar formats; they can also be used to digitally pan, tilt and zoom, and to create multi-window video.

Continuing along this same line, many applications are being designed and/or considered to help remove the human factor for everyday tasks. Increasingly, applications are being designed as object counters. For example, many states are looking at megapixel technology and system analytics in order to control traffic flow throughout their cities.

By analyzing the speed, automobile count and approaching traffic using special software, the traffic light system can be automatically controlled to improve the flow, and reduce the emission rate and gas consummation, as well as frustrated motorists.

This application would require only a 20-pixels-per-foot camera design to be successful.