When designing an IP surveillance system storage is an important part of the design, which if overlooked can cause a number of technical problems.
The move away from analogue cameras and the rapid increase in video resolutions and image quality for IP surveillance systems has created a need for more storage. A few years ago 1 Megapixel cameras were common but today, 3, 4 or above are normal specs for projects. Using codecs can help to decrease storage requirements by about 30% and therefore reduce costs.
A video compression codec helps reduce the video files and save storage and bandwidth for your recording system. Currently there are two main video compression formats, H.264 and motion JPEG. In this blog we will look at the new codec – H.265 – and see how it works.
How does the compression work?
H.265 works in much the same way as H.264, the camera sends I-frame/intra frame every second (or after a preset number of frames), this is a full image and is the starting point; after this the camera will send p-frames, that contain the differences between the I-frame and the frame it is trying to show.
The below picture shows what is sent if using JPEG: this format just sends the whole image for each frame, using large amounts of bandwidth.
With H.264 and H.265, frame 1 is the i-frame and contains the whole image, frame 2 is just the difference between itself and the i-frame; p-frames. Within the p-frames there is also data to allow the prediction of movement which lowers the data that is required to be sent. The following frames are again just the differences. Once sewn back together there is no drop in quality but the savings on bandwidth can be immense. The picture below shows how both H.264 and H.265 compression works.
This may provide savings in network bandwidth and storage costs but it requires higher performance in terms of processing power, for cameras, recording servers and client machines, as frames are encoded and decoded during data transmission.
How H.265 differs.
H.264 groups areas of the picture before sending, this can mean a large area is seen as moving and is sent across the network. H.265 allows these groups to be larger allowing less network traffic. These larger blocks also have greater motion vectors included; allowing for better prediction with less residual error.
This allows H.265 to send the same data as H.264 but use less bandwidth to do so. With the current trend in videos towards high resolution and frames per second the bandwidth savings become vital.