Video/IP/CCTV: How Many Megapixels Are Enough?

Nov. 19, 2012
Megapixel camera selection should be application-specific

More megapixels can equal better video systems. While it’s true that there are significant advantages to higher-megapixel cameras, they can also come with a downside. Weighing the benefits versus the challenges is key to achieving better video systems overall.

End users who migrate from standard VGA resolution to 1080p HD find that the upgrade provides clearer video and a better viewing experience. However, even that transition isn't without challenges, such as bandwidth issues and a need for a more robust, richer network environment to support HD. In fact, additional system requirements are one factor slowing down the migration to HD.

Megapixel cameras offering even more resolution are proliferating in the market and systems integrators and end-users are bombarded with marketing claims and new choices. The transition from VGA to HD is not happening overnight and if you go to even higher-megapixel-count cameras, the transition becomes a gigantic leap. A sweet spot in the market seems to be 1.3 and 3.0 megapixel cameras, which offer clearer, more detailed images along with real-time frame speed. This level of resolution appears poised to become the new industry standard over time.

Systems with higher-megapixel cameras have to control a lot more data and the associated costs can quickly become prohibitive. Decoding large video files also takes extra computing power. There are sensors on the market that can push the envelope in terms of resolution, but their usefulness will be limited and there is a tradeoff in frame rate.

The balancing act

It's important to balance additional resolution with the requirement for real-time motion. Many end-users want 30 frames per second (fps), especially in gaming applications, but higher-megapixel cameras may only provide one frame every second or so. Even if you scale back the frame rate to 10 or 15 fps for a general surveillance application, system requirements to accommodate the larger burden of video streams from high-megapixel-count cameras can be cost-prohibitive. The balance of frame rate to resolution and the impact on bandwidth and storage are all driven by the individual application.

There is unlikely to be a race to five or 10 megapixel cameras or higher because of the problem of transmitting images and limited bandwidth. Although compression is getting better, even sending 1080p video at 30 fps takes a toll on networks. With multiple cameras and larger systems, the impact is quickly compounded.

Another issue to consider with high-megapixel cameras is light sensitivity. Even though a camera may have more pixels, the sensor size is likely the same. With more pixels onto the same sized sensor, each pixel is smaller and therefore less sensitive to light. Another limiting factor in high-megapixel cameras is the availability of high-quality lenses with a sufficient transfer function to match a camera's capabilities.

Creating video systems with higher resolution is much more than simply choosing a megapixel camera. Camera choice is just one element in system design, and one that impacts all other elements and costs throughout the system. Choosing an HD camera must be combined with a network infrastructure that can efficiently move the images along. In the end, resolution is a system choice, not a camera choice.

When it comes to pixels, more is better, up to a point. But beyond this point (which is application dependent), the cost of accommodating higher resolution does not always justify the benefits. In the end, cameras that provide resolution higher than 1080p will mostly be used for specialized applications. A good rule of thumb is to put enough resolution where it is required, not automatically upgrade an entire system to the highest resolution simply because it is available.