Video Quality and Compression
Photo by Glenn Carstens-Peters on Unsplash
You must be familiar with compressing videos to save space. Large, high-quality videos take up a lot of space, and storing all those videos makes it a little crowded in the CCTV room and is also expensive. Losing a short 30-second video of the doorway may not be very damaging, but a theft due to insufficient storage space will have consequences. That is why video compression is a significant part of your surveillance system. Now, let’s take a look at the different types of video compression methods and the effect they have on video quality.
Video compression usually involves removing redundant and non-functional data in a video. There are four types of video compression techniques - H.264, MJPEG MPEG4, and H265 compression technology.
The first approach to compress a large file is to assess all the images in small groups and then remove the duplicates. This method is called H.264 compression, which is the most common technique used and is based on block-oriented motion-compensated integer DCT coding. In this compression technique, the frames are split into 8x8 or 16x16 pixel blocks. This algorithm usually predicts the next block with the help of a reference frame. The quality of the image is also not affected a lot in this compression algorithm.
The second one is also called motion JPEG, in which each frame of the video is evaluated and compressed. It is called MJPEG compression and treats each frame as an individual JPEG file. This compression technique does reduce the quality of the image when compared to H.264 compression.
But why does this happen? This is because the H.264 compression algorithm is a smarter algorithm. MJPEG studies individual frames and compresses them while H.264 groups frames, analyzes them, and then identifies duplicate frames that can be eliminated without actually reducing the quality.
See the differences yourself. The first image was compressed using MJPEG, while the second one was compressed using H.264.
Image from: https://ipvm.com/reports/h264-does-not-reduce-quality
Despite this, people use MJPEG widely because it has low latency and takes less recovery time in packet loss. Its low complexity also makes it popular despite its demerits of high bandwidth requirements and increased storage. The Nintendo Wii and PlayStation use this type of compression for videos.
The next compression technique is MPEG4. This method is now not used and has been replaced mostly bt H.264. Though it has higher compression rates and utilizes less bandwidth, high latency, and low robustness, users think twice before opting for this method.
Compression is significant when it comes to saving space and storing more videos. Compressing also makes it easier to send videos to remote appliances like your mobile phone. Storing and sending videos is often expensive. An uncompressed video that is around 100GB or more and takes $250 to store and transmit. Just think about the cost (to store and share) if the videos are in the size of terabytes! Compressing the videos proves to be a smart move here as well.
The latest compression technology used is H.265, which essentially offers the same level of picture quality as H.264, but with more efficient codec H.265 uses only half the bit rate of H.264. However, decoding H.265 streams require more computational power when viewing. When recording H.265, the NVR simply sees a smaller file to be stored.
The popularity of multi-megapixel cameras is the driving force for H265 compression technology.
Written By Haritha S.