Bit Rate in Video Surveillance with video StoragePlanning

Arindam Bhadra

What are Bit Rates?

Let’s start with “what is a bit, and “why do I need to know?” A bit is short for “binary digit”, the smallest unit of information in computing. It takes 8 bits to make a byte of information. “Bit rate” refers to the number of bits of data transferred in a file over a set length of time usually measured in number of “bits per second” or “bps”.

Constant bit rate (CBR) and variable bit rate (VBR) are the main types of bit rate encoding. Scene complexity can vary significantly over several hours of recorded video, and the bit rate you select for recording will have an effect on image quality, bandwidth consumption, and hard drive storage. A complex scene with moving action, such as traffic on a city street, or a scene with a lot of contrasting colors, will affect image quality and bandwidth consumption more than a less complex scene, such as an interior room with very little action or movement.

Most NVRs and IP cameras let you choose either constant or variable bit rates for recording video, and this is why you “need to know” the difference.

Constant Bit Rate (CBR)

With constant bit rate encoding, a fixed bit rate and bandwidth is used throughout the entire encoded video file. With a constant bit rate, image quality may fluctuate over the course of the video stream because some scenes are more difficult to render than others. In order for the bit rate to remain constant, the video may be encoded with fewer bits in some places or more bits in other places resulting in inconsistent image quality. Since bandwidth consumption with constant bit rates does not vary, the file size is limited and more predictable than with variable bit rates.

You will most commonly use CBR to restrict the data flow to keep network utilization as low as possible. If you have 10 cameras set to 8000K (8 megabits) on a 10/100 LAN, you are using 80% of your available bandwidth. With CBR, you can set that bit rate down to 5000K and your utilization will be around 50%.

Pre-planning your security video storage requirements is easier with constant bit rate because the amount of data being recorded never changes.

Variable Bit Rate (VBR)

With variable bit rate encoding, a changeable bit rate and bandwidth is used throughout the encoded video file. The variability of bit rates allows for video to be recorded at a lower bit rate when the scene on screen is less complex and at a higher bit rate when the scene is more complex. Complex scenes (such as moving traffic) require more data and greater bandwidth to maintain image quality than less complex scenes such as a wall or hallway with very little movement or action. With variable bit rates, the quality of video is higher and more consistent throughout the video stream compared to constant bit rates, yet the file size is less predictable.

Image quality is better with variable bit rates than with constant bit rates, yet pre-planning your security video storage requirements is more difficult because the bit rate changes and more complex scenes will require greater bandwidth and storage.

Here’s a Quick Look at How Constant and Variable Bit RatesCompare:

Constant Bit Rates

Variable Bit Rates

Variable video image quality

Consistent video image quality

File size is predictable because bit rate and bandwidth consumption is fixed

File size is unpredictable because bit rate and bandwidth consumption varies

Greater compatibility with most systems (compared to variable bit rate)

Less predictable compatibility (compared to constant bit rate)

When to use: When you need to limit file size and the quality of video is less important.

When to use: When consistent image quality is critical and predicting or limiting file size is less important.

The best of both worlds is when the device allows you to set VBR with a ‘Cap’ or maximum allowed bit rate.

Here is a handy ‘Quick Reference’ for setting a constant Bit Rate in bits per second.

Low Activity

Compression

Frame Rate

VGA/D1

720P/1.3MP

1080P/3MP

H.264

25~30

768K

2000K

3000K

15~20

512K

1500K

2000K

8~10

386K

1000K

1500K

2~5

256K

768K

1000K

MPEG4

25~30

1000K

3000K

5000K

15~20

768K

2000K

4000K

8~10

512K

1500K

3000K

2~5

386K

1000K

2000K

Normal Activity

Compression

Frame Rate

VGA/D1

720P/1.3MP

1080P/3MP

H.264

25~30

1000K

3000K

5000K

15~20

768K

2000K

4000K

8~10

512K

1500K

3000K

2~5

386K

1000K

2000K

MPEG4

25~30

1500K

4000K

6000K

15~20

1000K

3000K

5000K

8~10

768K

2000K

4000K

2~5

512K

1500K

3000K

High Activity or PTZ on Tour

Compression

Frame Rate

VGA/D1

720P/1.3MP

1080P/3MP

H.264

25~30

2000K

4000K

6000K

15~20

1500K

3000K

5000K

8~10

1000K

2000K

4000K

2~5

768K

1500K

3000K

MPEG4

25~30

3000K

6000K

8000K

15~20

2000K

4000K

6000K

8~10

1500K

3000K

4000K

2~5

1000K

2000K

3000K

In the bit rate charts above, you will see 1000K / 2000K etc. These figures can be loosely translated into ‘megabits’ per second.

1000K = 1Mb | 2000K = 2Mb and so on.

(In the computing world you would actually use 1024K = 1Mb and 2048K = 2Mb, but since most CCTV devices won’t allow those exact numbers, we just round them down to the closest thousand.) These figures are important to familiarize yourself with to manage your network load. For example – 1 camera running a high bitrate of 8000Kbps (8Mbps) is no problem on a 10/100 network. 10 cameras at that bit rate = 80Mbps. 80Mbps is 80% network utilization on a 10/100 LAN. This is enough to see visible slowdown on the network and may begin to cause problems.

Switch to a Gigabit LAN and that becomes 8% utilization. Always check the capabilities of the network you are installing on – this can save you from a lot of headaches. When using IP cameras, always use Gigabit routers and switches when possible such as this 8 Port POE Switch.  Also, make sure your NVR is connected to a gigabit switch. Plugging your NVR in to a 10/100 switch will limit your NVR to a 100Mb connection.

And finally – here is a “loose rule of thumb” for setting a bitrate:

[image width] x [image height] x [framerate] x [motion rank] x 0.07 = [desired bit rate]

Where the image width and height is expressed in pixels, and the motion rank is a number between 1 and 4. 1 being low motion, 2 being medium motion, and 4 being high motion (motion being the amount of image data that is changing between frames.).

So for instance, if we take a 1280×720 video at 24 FPS, with medium motion (movie with slow camera movements, not many scene changes…), the expected ideal bit rate would be:

1280 x 720 x 24 x 2 x 0.07 = 3,096,576 bps => approximately 3000 kbps, or 3MB

Remember – bit rates are not “universal” – different cams will give different results due to variations in encoding methods, hardware, and environmental conditions. Watch for artifacts like “ghosting” or “smearing” of moving objects.

Ghosting = when someone moving across the image may appear transparent, or may have a “ghost” following them. The “ghost” is not always transparent and may look like two people overlapped.

Smearing = when a moving object causes objects around it to change in appearance or starts to become pixilated.

Pixilated = When objects become unclear – may appear as a “smear” or slightly out of focus. In worst cases you will begin to see blocks of similar colors instead of the object itself.

Filed in: News

Recent Posts

Bookmark and Promote!

© 2017 ItVoice | Online IT Magazine India. All rights reserved.
Powered by IT Voice