Posts Tagged ‘ CIF’



Frame Rate vs Resolution in Security DVRs – Which is More Important

Written By:
Wednesday, May 21st, 2014

When configuring a DVR a very common question is what is the difference between frame rate and resolution and which is more important to have at a higher rate? Of course my answer is it depends… Both are very important but, depending on what you want to use the cameras for is which is more important than the other.

Now to begin you really need to understand what frame rate is. We will take a trip back in time to the 1800’s when cameras were first invented. They only took one picture at a time. The mechanism and film were designed that way. It was a physical limitation for decades. It was tedious and horribly long with the extra processing steps. You had to have fairly dangerous chemicals to actually develop the pictures. To take a picture, people had to stand as perfectly still as possible, because the camera’s shutter speed was so slow and the film need exposure times of ten minutes to an hour depending on the film. As film paper became more sensitive to light it reacted faster, so shutter speeds eventually had to be faster. I remember in the 1980’s the hot thing was Kodak High speed film. What set them apart from everybody else for a few years was Kodak figured out a way to get the film paper extremely sensitive. Coupled with cameras that had extremely fast shutter speeds Kodak figured out to how to take out the blur you may get when objects are moving. The pictures coming out of the Olympics that year were amazing there were no blur and the images extremely crisp. The faster shutter speed means a camera can take more pictures per second. Better film allowed for this by being more responsive to less light. A fact that is not well known by people is that the faster the shutter speed the less light comes in the iris which will create dim images.

Resolution is a separate aspect of the image compared to frame rate. CIF is 320 X 240 pixels. 2CIF is 720 X 240 pixels. 2CIF is a wider image than CIF depending on the need is if you would pick the wider shot or standard CIF. VGA is 640 X 480 pixels. D1 is 720 X 480 pixels. As image processing has improved the pixels have increased. With more pixels the colors become more vibrant and the images more clear. To better understand the clarity parts pixels are a very tiny piece of the puzzle known as your picture. The more pixels you can get the smaller the piece gets to represent that part of the picture. The down side is that this increases the need for better hardware to handle the increased load of imaging, which of course drives the price higher. That said the saying, “you get what you pay for”, could not be truer with DVR systems.

Resolution Comparisons

Now that we understand what both frame rate and resolution really are we can determine which is more important per the application. A few months ago I was helping a customer that had a DVR-EL4120ME. This DVR is capable of all the analog resolutions and the D1 at 7 frame rate but CIF at 30 frames per second. . I will say “John” was using this system to monitor a manufacturing process. The issue he was running into was that he was getting blurry images so he called his tech support GURU Daniel (that’s me). He goes on to describe what he is doing and the images he is getting. I logged into his system to see what he described. His process was happening so fast that he was not really getting an image of what he wanted, so I asked what is more important the resolution or frame rate. John said, “Gee I really don’t know. I thought resolution but with the images I’m getting I need help.” I told him no problem I have the knowledge to get it right. Instead of running the DVR at D1 and 7 frames per second, let’s adjust the system to CIF at 30 frames per second. POOF!!! Just like that the conveyor belt that was carrying some electronic device to be soldered could be seen. It was not about seeing exactly what was being soldered but making sure the robotic arm was hitting the part in time as it passed by. In minutes I was able to take John from not being that satisfied of a customer to being ecstatic about his equipment.

An example of the resolution being more important than frame rate is in monitoring your cash register. In many if not all retail environments keeping a close eye on the money is important. Typically people are not moving at warp speed so they do not look like a blur on camera. Giving way to frames per second while boosting resolution. Being able to make out the print on the one hundred, twenty, ten, five, and one dollar bills is important. People make mistakes on both sides of the register. The clerk gave the wrong change, the customer paid with the wrong bill etc. Now all of a sudden being able to make out the print is a big deal, so you know without a doubt who gave what to who.

Cash register

I can tell you having a customer have a meltdown at the main register is really not good so, as a manager you need a way to quickly defuse the situation before it carries over to other customers. Having your DVR configured to the highest resolution lets you see the 10, the 20, the 50, and 100 dollar bills. You can go back to where the DVR is setup, review the footage and know with confidence what happened or who made the mistake. Also, you can show the customer in real time “who” made the mistake.

Now you know the difference between resolution to frame rate and in which applications makes either feature more important. Should you run into a scenario when you are not able to compromise, IP cameras typically are the way to go. They support the mega pixel resolutions which are considerable higher than analog as well as 30 frame rate per second.

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail

4CIF Resolution

Written By:
Sunday, April 22nd, 2012

4CIF ResolutionUsing the 4CIF Resolution to Open the Eyes of Surveillance Cameras

The 4CIF Resolution format is common enough that it warrants some discussion. Image resolution becomes very important when discussing DVR compatibility. Most processors really don’t perform well when it comes to recording in a real time resolution. However, there are some units that are able to record in 4CIF on all channels in real time. This is an impressive accomplishment. Many DVR units only record at the CIF resolution at around 30 frames per second.

Many people are already familiar with resolution because of digital camera technology. Resolution essentially measures the quantifiable amount of detail in an image. This is generally expressed in terms of pixels when talking about digital imaging. In a weird twist of fate, the term resolution is actually wrong in this situation. Industrial guidelines published in the United States and Japan state that the term resolution isn’t supposed to be associated with digital camera pixel counts. Few people would ever write the term “number of recorded pixels” as a label. Nevertheless, the translation of the Guideline of the Camera and Imaging Products Association states that it is preferable to resolution. While one might never come across CCTV that actually uses that name, it’s important to keep in mind just in case.

On the other hand, most people are probably unaware of what the individual acronyms stand for. CIF means Common Intermediate Format. Some texts might refer to it as the Full Common Intermediate Format and abbreviate it as FCIF. CIF was supposed to standardized the vertical and horizontal pixel configurations in video signals transmitted to YCbCr color space sequencing devices. The idea was that one could easily convert these signals between the NTSC and PAL formats. PAL televisions are popular throughout many European, African and Asian nations. Consumers in the United States and Japan are far more familiar with the NTSC format.

These standards were first proposed in H.261. That document set the gears in motion that would ultimately create the 4CIF resolution. It was ratified in November 1988, and it was the first member of the H.26x series of video code standards that the Video Coding Experts Group laid out. People that have been working in the surveillance industry for some time might also remember H.261 as being the first set of video codec standards that had any real practical use.

It was originally set up to enable the transmission of video data over ISDN lines. Many modern security installations aren’t organized in this fashion. However, the 4CIF resolution is still a useful counting metric. The algorithm used to code the video was designed to operate at bit rates that were somewhere between 40 kbit/s and 2 Mbit/s. It should be noted that ISDN data rates are measured in multiples of 64 kbit/s.

One of the more interesting aspects of H.261 encoding involves a trick where backwards-compatibility functions can be used to send still pictures. Graphics that are formatted for 704 x 576 luma transmission go through just fine. In 1993, the standard added a method for transmitting 352 x 288 chroma resolution images. These measurements are technically outside of the usual definition of the 4CIF resolution. However, the 4CIF resolution is defined as a 704 x 576 matrix. Some users have surely messed around with that little detail.

The regular Common Intermediate Format resolution is defined as CIF at 352 x 288 pixels. Other resolutions in the standard are expressed as some measurement in reference to this base. The Sub Quarter CIF (SQCIF) format measures in at 128 x 96 pixels. Cameras that claim to use the Quarter CIF (QCIF) resolution should have a video resolution of 176 x 144 pixels. SCIF is measured at 256 x 192 pixels, but these small resolutions are generally rare in the CCTV industry.

Thus the 4CIF resolution is considerably larger than the standard CIF resolution of 352 x 288 pixels. There are even large options for those that can handle them. The 16CIF mode is effectively double the 4CIF resolution and weighs in at 1408 × 1152. Needless to say, most equipment doesn’t enjoy real time functions in that size.

Many CCTV experts couldn’t care less about this sort of nitty gritty technical discussion. They might be interested in the fact that 4CIF Resolutionsome of their job can be done with software, however. There is an algorithm distributed under the LGPL license that can work as an H.261 encoder and decoder. The libavcodec contains numerous other codecs, and might be a useful tool for many surveillance engineers that want to interface different types of hardware. One could also probably use it to convert older video into an easy to read format. This particular algorithm enjoys support from the MPlayer multimedia player as well as the free VLC media player. The FFmpeg and ffdshow projects also interface with the H.261 standard. Even if vacation photos don’t show up in the 4CIF resolution, it might be worth it to grab a few of these pieces of software. They can make a CCTV run more smoothly and eliminate incompatibility headaches.

That being said, these problems won’t come up if all of the hardware was purchased together. The Security Camera King features several devices that natively support the 4CIF resolution. Linking devices that weren’t intended to work together usually causes compatibility related errors. The King has customers covered. Cameras and DVR machines that can function in the 4CIF resolution format will undoubtedly do so when the King has given them his seal of approval.

The 4CIF resolution is sometimes called the 4SIF format. This can be very confusing, and isn’t entirely accurate. The Source Input Format (SIF) is essentially identical to the CIF, but it was taken from the MPEG-1 format as opposed to an International Telecommunication Union document. Regular SIF formats on 525-line NTSC display terminals will exhibit a resolution of 352 x 240. Things are a little different on 625-line PAL displays, where the format is practically indistinguishable from the CIF resolution. This means that the 525-line version of 4SIF isn’t the same as the 4CIF resolution, but the 625-line version of it essentially is. The SIF and 4SIF standards are pretty common in the world of video conferencing. Some users like to interface CCTV machines with video conferencing terminals.

This can be particularly useful for anyone trying to quickly share evidence. Therefore, installers will probably want to keep those tips in mind. However, people that work with standard CCTV setups can usually ignore the additional advice. Most people won’t have much of a need to interface PAL and NTSC equipment together. Digital formats can sometimes handle this, but plugging an NTSC camera into a PAL monitor is a recipe for disaster. The inverse is also true.

Some people have probably messed around with the PAL and NTSC settings on regular DVD players. When one does that, they quickly find out that the incorrect setting causes unpredictable results. Since the scan lines don’t match up, it often causes the video display to scroll up the back of the screen. This looks strange and might damage some terminals.

While Video CD isn’t as popular a format for many people as DVD is, some surveillance engineers have probably worked with it. It’s important to note that these designers of these devices choose the SIF format for a reason. The 4CIF resolution isn’t directly compatible with the VCD format. Some programmers can probably work some magic to get them to work together. However, most people in the surveillance field wouldn’t care to convert 4CIF resolution encoded video to the VCD format anyways.

SIF is displayed at 352 x 240 on 525-line displays and at 352 x 288 pixels on 625-line displays. This puts it at half the vertical display resolution of NTSC video. Likewise, it is half the horizontal display resolution of PAL video. Likewise, the format displays video at around half the resolution of a standard analog VHS tape. VHS tapes generally work around 330 x 480 on NTSC equipment and 330 x 576 on PAL devices. This information is important to keep in mind when one wants to use the 4CIF resolution with something that it wasn’t intended for.

Those that are interested in interfacing the 4CIF resolution with other formats might be interested in reading more about the White 4CIF ResolutionBook CD standard. White Book defines the modern VCD and SVCD formats. The file system is supposed to be compliant with ISO 9660. The maximum length is usually around 74 minutes, which may or may not make it unsuitable for many closed circuit television applications. The audio and video are both compressed with the MPEG-1 format, though the SVCD format relies on the MPEG-2 algorithm for video compression. Since the audio is transmitted as a two-channel stream, White Book videodiscs are compatible with stereo recording equipment.

Most DVR machines can record at the maximum frame rate in CIF resolution. Recording in the 4CIF resolution usually involves a reduction in frame rate. However, it provides an image that is literally four times as large. Therefore, one can get far more detail out of this sort of a picture. That extra detail is often necessary. This is especially true when large areas are being monitored. Detail can make up images that lack in other types of depth. However, there are other situations where one would primarily care about the way in which the images themselves recorded. Frame rate could be important. Streaming video to other people usually requires a pretty fast frame rate. This is necessary to avoid making the stream look choppy. Of course, frame rates in streams could end up being choppy anyways. The transmission speed has a lot to do with this.

When transmission speeds decrease, frame rates probably won’t mean much. However, one also wouldn’t want to increase the size to full 4CIF resolution. This would simply make the problem worse by making the file size even larger. That’s why it might be best to stick to equipment that was designed to interface with the 4CIF resolution directly. A DVR machine that can accept a 4CIF resolution feed and record it without reducing the frame rate is a thing of beauty.

Once in a while, people that use the 4CIF resolution will come across DCIF equipment. The Double CIF standard was supposed to be a compromise between CIF and the 4CIF resolution. It’s supposed to be more balanced in terms of vertical and horizontal display resolution. It was also supposed to be more compatible with common CCTV equipment. To that end it displays over 480 scan lines but tops out around 560. The pixel measure is exactly double CIF at 528 x 384. The 1:1.375 image aspect ratio is said to be closer to the standard of 4:3 while showing square pixels. CIF pixels are generally not truly square. Instead, they have a native image aspect ratio of somewhere around 1.222:1. A 1.2:1 aspect ratio was one a common standard for 525-line displays. That’s where the strange measurement probably came from.

Many modern televisions display square pixels. This is also true of computer monitors. Since many people use modern equipment to display the 4CIF resolution, this can be a problem. Some 4CIF resolution equipment ends up looking somewhat stretched.

This can be avoided by horizontally rescaling the 4CIF resolution rasters by approximately 109% to give a standard 4:3 aspect ratio. Standard CIF displays that transmit a video stream at 352 x 288 would thus end up displayed as 384 x 288 square pixels. It should be noted that some of this discrepancy might be explained by the fact that the H.261 outlines what are now considered to be legacy formats. Later standardization efforts have tried to refine the design and improve compression rates. Many engineers consider the H.261 outline to be obsolete. However, there are plenty of video conferencing terminals that continue to cling to it. It is offered as a backward-compatibility mode in many of these. Certain Internet video streams also use the format. The 4CIF resolution won’t be going away any time soon.

 

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail

Security DVR Recording Resolutions

Written By:
Friday, March 19th, 2010

You’re in the market for a security surveillance system for your home or your business. What is the one most important component of that system? The Digital Video Recorder, or DVR, is the single most important component. And the most important aspect of choosing a DVR is the recording resolution.

In manufacturing a DVR, the recording resolution is a balancing act between being able to actually see what’s being recorded and having the storage space to store the video. The better the resolution, the better the images look, but the more space it takes to store it, and the beefier the processor it takes to record it. Manufacturers try to offer that ideal balance between being able to actually have usable recorded footage for evidence of an event if needed, being able to process the enormous amounts of data involved while recording it at a usable resolution, and not running out of recording room to store events. How the manufacturers succeed at this is what you need to evaluate when shopping for DVRs. Budget will be a huge consideration here, since the units that juggle these factors in the most optimum way are, of course, the most expensive. Putting together the optimum system which takes into account all your needs, including your budget, and gives you recordings from all DVRs that can actually be used as evidence takes a bit of knowledge and the help of a qualified reputable professional.

When referring to recording resolution the industry usually uses either a variant of CIF, which is 360 x 240 pixel resolution, or D1, which is 720 x 480 pixel resolution. There is a distinction that needs to be addressed here. Each DVR has a recording resolution and a live video resolution. The live video is always going to be clearer, so when you are evaluating DVR units you really need to know how the recorded video will look, since that will be your actionable evidence in the event of a security problem. D1 is currently the highest resolution used for recording, but most standalone DVRs are not able to record in D1 in realtime, which is 30 frames per second (fps) on all the channels at the same time, though some of the new units coming out are starting to improve on that.

The strength and capacity of the processor that is running the DVR will determine how high a resolution can be set for recording. The memory in the unit is also a big factor. Most processors can’t yet record at a realtime resolution, though some of the newer units that are beginning to show up in the marketplace are hitting 4CIF recording in real time on all channels, which is a 704 x 480 pixel resolution. The current average for a DVR is a CIF recording resolution, which is 360 x 240 pixels in real time at 30 fps.

When looking at the specifications, the three things you need to consider together are the recording resolution, the frame rate, and the recording speed. Of the three, the recording resolution carries the most weight in a carefully considered purchasing decision. If there is a security event, you want to be able to clearly identify the face of the person involved, and if the resolution doesn’t allow this, it’s useless. A 640 x 480 recording resolution is the minimum recording resolution which will easily allow this. 720 x 480 has the same quality but in a wider screened aspect ratio to allow viewing on the typical widescreen computer or tv monitor. 1600 x 1200 is ideal for viewing from an actionable standpoint, but the speed, storage and processing needed to maintain this recording resolution make the cost prohibitive for all but the most deep pocket organizations.

When putting together a budget system, the best way to approach it is to choose a better resolution for key points of security, ie near the front door and the cash register, while choosing lesser resolutions for areas which would be more likely to be monitored in realtime but where recording resolution is a bit less critical, and would simply be support footage when combined with the higher resolution footage from the other areas, not an ideal approach, but one which would allow security on a smaller budget.

Whatever your security needs, the help of a qualified and reputable security monitoring specialist will shortcut your efforts enormously. Juggling all the factors involved in the purchase, including the recording resolution needed for actionable evidence, is key to the overall success of your security efforts.

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail

Basics of Facial Capture and Facial Recognition

Written By:
Sunday, January 31st, 2010

It is likely that if you get a security camera system, that you will want to be able to identify an intruder that has been captured on security camera footage. This process is called facial capture. Some people also call it facial recognition although they are really two different processes. Facial capture means to get enough usable details to be able to identify the person of interest by comparing facial features. It doesn’t mean the security cameras will identify the person, only that enough detail has been captured that he can be identified. Facial recognition actually refers to analytic software that can compare and match captured images of a persons face against another in a database. There are many interesting applications for such a process, but that is an entirely separate article.

Getting proper facial capture is an important process to understand because what good are your security cameras if you cannot identify the person of interest in the video? Proper identification requires at least 25% of the video image to be occupied by the face. This means that you really need to use a dedicated camera for this process. Begin the installation by finding an ideal spot to capture the face. Usually an entrance or exit would be ideal, but maybe there is a room entrance to something valuable. Once you have found the perfect spot to capture the face details, decide on the type of camera you will use for this process. It is best to use a good quality day/night security camera with a vari-focal lens rather than an infrared camera. Infrared cameras will be taking video in black and white in low light and this could make identifying the intruder more difficult. Since you will be using a good low light camera, make sure there is at least a little bit of ambient lighting. A night light, motion light or something of the like will be fine. Mount the camera close enough to the capture point so that at least 25% of the video (preferably much more) is the subjects face. Make sure to accommodate for people of different heights. Next, mount a second camera in a corner, or somewhere else in the room to provide you with a good overview. Now you have one cameras that is used to only capture facial details and a second camera that is providing the general overview of what is going on in the room. Make sure that the channel that the facial capture camera is on is set to record at the highest resolution possible. A D1 resolution image will be 4 times larger than a CIF resolution image and therefore will provide you with 4 times more detail. Use the higher resolution setting even if the frame rate has to be reduced. The overview camera can be set at CIF resolution rather than D1.

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail

How To Set Up Your DVR For Recording

Written By:
Tuesday, November 17th, 2009

Keep in mind that all DVR’s have different configurations and terminology. I will explain how to properly set up the Elite, Elite-Mini-HD, and Ultimate DVR’s for recording while conserving  maximum storage. Most DVR’s will not have as many options for recording configuration, but this should cover just about every setting you will find.

In the Elite, Ultimate and Elite-Mini-HD series DVR, every camera has its own configuration. Unlike most DVR’s, you can set each camera differently, or they can be all the same. Most other DVR’s have only one setting that applies to all the cameras at once. Let’s discuss each setting individually.

Resolution:

I recommend setting overview cameras to CIF recording (360×240). Set any cameras that need extra detail to D1 (704×480) recording. This would apply to cameras that are used for license plate capture, cash register drawer capture or facial capture at entrances or exits.

Frame Rate:

I would set up the overview cameras to 15fps. You will notice very little difference between 15fps and 30fps. Since these are overview cameras, there is really no need for 30fps. I would set detail capture cameras to 30fps.

Quality level:

Set all cameras to the highest quality level.

Bit Rate:

Set the Bit Rate to match the available bandwidth. I recommend using no more that 40% of the bandwidth so if you know you have 2 mbps up bandwidth available, and you have 8 cameras, I would set the bit rate to 100kbps. Also, set the bit rate to VBR rather than CBR. VBR allows for variable bit rate which gives the DVR the ability to throttle the rate according to available bandwidth. CBR is for constant bit rate.

Recording:

Set the DVR to record motion only.

Pre and Post Recording:

Set the prerecording to 10 seconds and the post recording to 10 seconds. This will provide you with recordings that have 10 seconds before and 10 seconds after the motion event.

Sensitivity:

This will need to be set individually on a per camera basis. Set the sensitivity so that the DVR is only triggered to record by objects you want to trigger. For example, you may not want a bird, dog or cat to set of the recording, but you may want a person to trigger recording. This is done using the sensitivity level.

Mask:

Make sure to mask out any areas of motion that you do not want to trigger recording. For example, you probably would not want a flag on a pole or a moving tree to trigger the motion recording. Just mask those areas out on a per camera basis.

If you configure everything as stated above, you should be able to maximize the storage ability of your hard drive. Keep in mind that these are only suggestions. There can be many deviations from these suggestions depending on your preferences and needs. Also, if you have bandwidth issues to deal with and do not have a sub stream available, then you may need to lower some of these settings.

The Elite-Mini-HD and the Ultimate Series DVRs both have sub streams available, so you can leave the recording settings to optimal.

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail