Video Compression Not All are Created Equal
Video compression makes it possible to reduce the size of digital video files while maintaining a useful level of image quality, thus allowing for more efficient management of file sizes and bandwidth requirements. But not all compression standards are created equal. In particular, the new H.264 standard is rewriting the rules of what’s possible in today’s world of video compression and streaming network transmission.
Digital video, particularly high resolution megapixel digital video requires a lot of computer bandwidth as it travels across a network. It also requires a greater amount of data storage capacity to hold and store such files. Compression is a tool to manage video data usage and minimize bandwidth/storage needs. Some of the more historically popular standards for video compression are:
• Motion JPEG (MJPEG) compression requires each individual video frame is encoded as a separate JPEG image. Any single frame of video can be captured, preserved or printed out as evidence. However, the video takes up too much bandwidth to be practical for most streaming video applications.
• Motion JPEG2000 compression is an update to MJPEG that uses wavelet-based image compression for improved performance over JPEG while adding features such as scalability and the ability to handle a large range of effective bit rates. JPEG2000 offers about a 20 percent gain over standard JPEG compression, but requires more computing power and memory to accomplish.
• MPEG-4 compression does not encode each frame separately. Some frames
(I-frames) contain complete data, but between the I-frames are “predictive” frames that include only partial information, just what has changed from the I-frame. In the case of a scene that doesn't have much motion – for example, a camera watching an empty hallway at night – MPEG-4 uses less bandwidth because little changes from frame to frame.
Some cameras and systems use a combination of compression standards – dual streams of video. MPEG-4 supplies streaming video from a camera across the network, but Motion JPEG is also used to supply still images, sometimes recorded locally and used in case of an incident.
The newest video compression standard is H.264, which can save bandwidth and storage by compressing even the highest megapixel images to manageable sizes. However, the encoding process is more complex and more advanced processing is required to compress the video. In the beginning this need for additional computing power complicated the adoption of the H.264 standard. Now however, many companies have developed dedicated H.264 compression chipsets that handle all the “heavy lifting” for the device that uses them. This has positioned H.264 as the standard to deliver the perfect balance of low bandwidth, high quality images and high frame rates. In fact, H.264 compression technology works so well that many big companies outside the security industry are quickly implementing H.264 technology, including Apple, Intel, Blu-ray, Sony and even satellite providers such as Direct TV and Dish Network.
Because H.264 provides superb compression, in most security applications it more than doubles the record times compared to previously popular compression methods. Here is a comparison of H.264 to other compression technologies, using a 160GB hard drive, recording video at 30 frames per second, at a resolution setting of 720 x 480:
• H.264 records 77 Hours
• MPEG-4 records 44 Hours
• JPEG2000 records 13 Hours
The ability to record longer periods of time on the same-sized hard drive can save your customers money, and greatly increase the amount of video evidence they can archive.
Although other compression methods may provide good picture quality at real-time frame rates, they require more hard drive space or network bandwidth. Similar to MPEG-4, H.264 uses predictive technology to drastically reduce the amount of redundant video recorded. In simple terms, consider an unoccupied room with no movement; instead of constantly re-recording, the same scene over and over, H.264 will simply use previously recorded images until the scene changes. When someone walks into the room, H.264 will record that person and continue to use the previously recorded background. H.264 will sample up to 32 frames of video to ensure complete accuracy and quality, whereas other compression methods would use only 1 or 2 frames as a reference. In essence, H.264 combines the most desirable benefits of many previous generations of compression standards.
In addition, H.264 can bring together high quality and low memory sizes to allow more seamless presentations of video when transmitted. Your customers will find watching real-time H.264 to be a smoother experience where they don’t even realize that the video they’re watching has been compressed. This explains why so many broadcast companies around the globe have adopted it. However, it is also important to note that while H.264 compression is a step forward in video technology for all industries, it does typically require either a dedicated chip for handling the compression (such as in Apple’s iPhone), or the allocation of more computing resources to deal with the real-time video streams when using a conventional processor.
Digital video, particularly high resolution megapixel digital video requires a lot of computer bandwidth as it travels across a network. It also requires a greater amount of data storage capacity to hold and store such files. Compression is a tool to manage video data usage and minimize bandwidth/storage needs. Some of the more historically popular standards for video compression are:
• Motion JPEG (MJPEG) compression requires each individual video frame is encoded as a separate JPEG image. Any single frame of video can be captured, preserved or printed out as evidence. However, the video takes up too much bandwidth to be practical for most streaming video applications.
• Motion JPEG2000 compression is an update to MJPEG that uses wavelet-based image compression for improved performance over JPEG while adding features such as scalability and the ability to handle a large range of effective bit rates. JPEG2000 offers about a 20 percent gain over standard JPEG compression, but requires more computing power and memory to accomplish.
• MPEG-4 compression does not encode each frame separately. Some frames
(I-frames) contain complete data, but between the I-frames are “predictive” frames that include only partial information, just what has changed from the I-frame. In the case of a scene that doesn't have much motion – for example, a camera watching an empty hallway at night – MPEG-4 uses less bandwidth because little changes from frame to frame.
Some cameras and systems use a combination of compression standards – dual streams of video. MPEG-4 supplies streaming video from a camera across the network, but Motion JPEG is also used to supply still images, sometimes recorded locally and used in case of an incident.
The newest video compression standard is H.264, which can save bandwidth and storage by compressing even the highest megapixel images to manageable sizes. However, the encoding process is more complex and more advanced processing is required to compress the video. In the beginning this need for additional computing power complicated the adoption of the H.264 standard. Now however, many companies have developed dedicated H.264 compression chipsets that handle all the “heavy lifting” for the device that uses them. This has positioned H.264 as the standard to deliver the perfect balance of low bandwidth, high quality images and high frame rates. In fact, H.264 compression technology works so well that many big companies outside the security industry are quickly implementing H.264 technology, including Apple, Intel, Blu-ray, Sony and even satellite providers such as Direct TV and Dish Network.
Because H.264 provides superb compression, in most security applications it more than doubles the record times compared to previously popular compression methods. Here is a comparison of H.264 to other compression technologies, using a 160GB hard drive, recording video at 30 frames per second, at a resolution setting of 720 x 480:
• H.264 records 77 Hours
• MPEG-4 records 44 Hours
• JPEG2000 records 13 Hours
The ability to record longer periods of time on the same-sized hard drive can save your customers money, and greatly increase the amount of video evidence they can archive.
Although other compression methods may provide good picture quality at real-time frame rates, they require more hard drive space or network bandwidth. Similar to MPEG-4, H.264 uses predictive technology to drastically reduce the amount of redundant video recorded. In simple terms, consider an unoccupied room with no movement; instead of constantly re-recording, the same scene over and over, H.264 will simply use previously recorded images until the scene changes. When someone walks into the room, H.264 will record that person and continue to use the previously recorded background. H.264 will sample up to 32 frames of video to ensure complete accuracy and quality, whereas other compression methods would use only 1 or 2 frames as a reference. In essence, H.264 combines the most desirable benefits of many previous generations of compression standards.
In addition, H.264 can bring together high quality and low memory sizes to allow more seamless presentations of video when transmitted. Your customers will find watching real-time H.264 to be a smoother experience where they don’t even realize that the video they’re watching has been compressed. This explains why so many broadcast companies around the globe have adopted it. However, it is also important to note that while H.264 compression is a step forward in video technology for all industries, it does typically require either a dedicated chip for handling the compression (such as in Apple’s iPhone), or the allocation of more computing resources to deal with the real-time video streams when using a conventional processor.
Looking for a reprint of this article?
From high-res PDFs to custom plaques, order your copy today!
