Q&A: Sony's Mike McCann on IP video surveillance

In advance of our Aug. 6, 2010, webinar on IP video technology with Sony, SecurityInfoWatch.com caught up with Mike McCann, the manager of solutions engineering at Sony Security Systems Group (a unit of Sony Electronics Inc.) to talk about network video surveillance trends. McCann brings a deep knowledge of IP video from inside one of the industry's largest manufacturers, and his prior background working at an IT systems integrator has kept him on top of IT and network knowledge. To register for the free webinar on Thursday, Aug. 6, visit our webinar registration page. Attendees will get a chance to participate in a live Q&A with McCann as part of the program. The webinar will be archived after the live event, and you can use that same registration link to view the archived webinar.

SIW: HD and megapixel seem to be competing for the attention of buyers today. What do you see as the difference? Why would one have an edge over the other?

McCann: HD and Megapixel are often used as interchangeable terms but there are some notable difference. First, HD is a video format standard with both 720p and 1080p resolutions in the 16:9 aspect ratio (1280x720 and 1920x1080). The term megapixel does not describe a video format but simply counts the number of pixels on the sensor. Typically, Megapixel is a 4:3 aspect ratio with resolution settings such as 1280x1024 or 1920x1440. In some cases, a camera could have both an HD mode and a megapixel mode. Of course, both HD and megapixel network cameras use a variety of video compression technologies. Instead of dwelling on the term HD or megapixel, I think security professionals need to educate themselves on the overall performance of a camera. The quality of the lens, the performance of the sensor technology, the power of the DSP (digital signal processing) features and the encoding performance of a camera must all come together to deliver high quality network video.

Integrators have been struggling with the concepts of variable bit rate and constant bit rate in terms of video data streams. What is your take on these two streaming methodologies?

A variable bitrate cameras can offer a low bandwidth usage when there is no activity but can spike up unpredictably high when there is motion in the scene or when the video signal is amplified in low light conditions. For a small camera system on a wired LAN this may not be a big problem but in large systems or wireless systems unpredictable variable bitrates could become a problem for reliable connectivity and storage calculations. For example, a variable bitrate in a 720p camera could swing from 2Mbps to 6Mbps depending on the amount of activity and lighting conditions.

A constant bitrate will have a more predictable impact on the network bandwidth usage. However, system administrators need to be aware of what bitrate settings are required to achieve acceptable video quality. For example, A 720p constant bitrate camera could be set at 3Mbps and remain constant 24 hours per day to delivery very good video image quality. However, a constant bitrate of 512kbps would not be recommended for 720p HD video.

The camera lens didn't get much attention for a long time, other than making sure you picked a lens that gave the right zoom or field of view. But now, the lens seems important again in this age of high resolution video. Why is that?

Anyone with a background in photography or professional video knows the importance of lenses and lighting. Too often, analog CCTV cameras are sold with inexpensive lenses which can degrade image quality but often pass as "good enough". In HD and megapixel cameras, customer expectations are higher and lens quality is much more critical. HD and megapixel camera lenses require both high quality optics and precise focusing. Look for lenses that meet or exceed the resolution of your camera. Selecting cameras with an "Auto Back Focus" feature will also provide focus precision while saving valuable time and money.

Sony has been selling image sensors for years to other manufacturers, but what has been happening in the area of digital signal processing from Sony?

Sony is very well known as a supplier of CCD and CMOS sensors in the security camera industry. However, Sony DSP technology in Sony brand cameras is often taken for granted. Wide dynamic range, digital noise reduction and true color reproduction are accomplished by integration of the sensor and DSP. Sony Exmor CMOS sensors found in Sony Gen5 IP cameras delivery new levels of sensor and DSP performance in 720p/1.3MP and 1080p/3MP models.

How important is pixel density when it comes to image quality?

The spec sheet will tell you how many pixels are on the CCD or CMOS sensor but it is very important to understand how to set up the lens field-of-view to meet customer expectations. Many people assume they can simply use an "ultra wide angle" lens when using an HD or megapixel camera. However, this can often lead to disappointment when they realize the "pixels per foot" or "pixel density" in the scene becomes very low as you move away from the camera. While pixels per foot are important, let's not forget that there are many other variables in getting good video. It takes more than just pixels on a sensor to make a good camera. In addition to pixel density, cameras need to perform in low light and the wide dynamic situations without washing out and without excessive AGC signal noise and slow shutter.

[To hear more of McCann's thoughts on IP video surveillance, register to attend SecurityInfoWatch.com's Aug. 6, 2010, webinar.]