|
|
|
Alaska Business License # 712765
|
The common misconception of digital imaging devices and the images that result from them is that digital is better than analog, unfortunately no device has the ability to digitally detect, measure, or record any nature or essence of the real world. The ability to capture, measure and record is done by electrical devices that convert monitored change, movement, sound, light, temperature or any other measurable essence of the universe into analogous electrical frequencies or voltage values. These sensors then send out analogues electrical signals to other devices that either use the signal as is or converts it into some other useful form. To create digital images the signal is sent to an analog-to-digital converter that changes electrical voltage into digital numbers which can be stored on a computer. There are currently two types of analog sensors available to convert light into analogues electrical signals. They are the Charged Couple Device (CCD) imaging sensor and the Complementary Metal Oxide Semiconductor (CMOS) imaging sensor. Both are different technologies and each has unique strengths and weaknesses. However, both imaging sensors do the same job of converting light into electric charge and sending a signal to another converter that changes the signal into a digital number that can be stored on a computer. The analog-to-digital conversion causes the ability to store and manipulate an image on a computer. It is the analog-to digital converter causing digital imaging to be better than analog imaging, but this conversion does not necessarily mean the resulting picture is better than what can be provided by film or analog video technology.
While digital imaging does require a continuous electrical stream having variations to start the digitizing process, once this signal is converted into a digital data stream the original audio and image data can be preserved indefinitely and copied or relayed through relay transmitter repeaters repeatedly without deterioration or loss of integrity. The shorter the time interval between the original analog capture and its conversion to digital and the more data recorded by the conversion process, the more the digital encoding reflects the original analog signal. Numbers cannot fade, color shift, or deteriorate like dyes used in film and the decoding of numbers does not require calibration, amplification to copy. The exact replication of numbers back into sound or image is not distorted by extraneous noise that come from unwanted background patterns of mechanical, electrical, or magnetic vibrations that come from not only the material the analog signal is stored on but from nearby inferring signals from other sources, from faulty bias, from poorly fitting or dirty contacts (stylus tip or tape head), and from power spikes. Although numbers do not fade or change each generation until there is nothing worthwhile to copy or replicate like analog, digital imaging has its own limitations and restrictions. The most inconvenient limitation of digital imaging and specifically with digital video is the unique complications in that converting the voltage value signal to numbers results in an unmanageable amount of numbers so a compression/decompression scheme or codec is used to create the data file recorded on the digital video tape, digital disc, or memory card. The conversion of motion and recorded time sequence to digital video is further restricted in the data has to flow and be encoded and decoded at some expected data stream rate. While there is data capability difference between formats and codecs, the desired results is to reproduce an accurate and pleasing picture and the basic quality of the data stream written to tape or in a file is the same.
The data rate, sample ratio, and compression ratio simple describe how much information is packed into numbers and how many result to represent a complete image numbers. Higher data rates mean either more numbers can be transferred or larger more accurate numerical structures can be transferred in an expected period. The sample ratio and compression ratio describe how structured or accurate the data is. Video pictures are composed of four elements that must be stored and reproduced properly. These elements are described in a notation comparison called the sample ratio. The first number in the comparison describes how often the luminance ‘Black & White’ (Y) portion of the analogous signal is sampled, the second term describes how often the ‘Red’ color (R-Y) portion of the analogous signal is sampled, and the third term describes how often the ‘Blue’ color (B-Y) portion of the analogous signal is sampled. The ‘Green’ color is calculated from the luminous and red and blue values. A 4:1:1 sample ratio is a step down from a 4:2:2 sample ratio; however, the resulting image difference between the two ratios is hard to notice. The difference is most often noticed by the video editor desiring to edit into a sequence a text, graphics, and other special effects that call for sharp colors that can’t be accurately described using a 4:1:1 sample rate. Compression is used to reduce data flow, storage space, and tape consumption. However, the less you compress the data, the better the quality. No compression results in better image quality than 2:1 and 3.3:1 compression. Both 2:1 and 3.3:1 compression ratios result in essentially providing image quality that is hard to differentiate from no compression. The more compression used, the more artifacts become visible and a 10:1 compression can be achieved before loss of image quality becomes noticeable. A 5:1 compression results in good image quality and is used primarily to make digital video and digital video tape economically affordable to the typical consumer. The price for not compressing is high. High data rates above 142 Mbps require top-of-the-line circuitry to handle the data stream and lots of tape footage and disc space to store all the digits generated. All imaging but especially digital video imaging is dependant on making advantage of the characteristics of the human perception system. The converting of an analogous video signal from an imaging sensor into digital sequence of numbers that describes the picture or a frame of one picture frame of many in a sequence of video involves many complicated elements of process. A concurrent compression and conversion process begins the moment the CCD or CMOS imaging sensor starts measuring. Not only does the design specification of the imaging sensor determine how much it can collect and measure, there is a rate and control logic circuit controlling the sample rate and data flow inputted from the sensor into the analog-to digital converter. The analog-to-digital converter then uses a conversion and compression algorithms (codecs) to encode numbers to be put onto tape, disc, or computer memory. Codecs are compressing-decompressing pairs as whatever is encoded has to be deencoded to be seen as an image and they are designed to be either ‘lossless’ and ‘lossy’. Lossless compression is perhaps more accurately described as an encoder as there is minimal compression that is typically less than 2:1. The decompressed output of the resulting from a lossless date file matches the original data inputted from the originating source of the data or analogous stream. Lossy compression does not preserve all the data that is inputted from the originating data stream. It reduces size of the data being saved by discarding information that is not noticeable and is excess because it is expected the human ear and eye will not detect its absence. Although, lossless compressors are useful for digital scanning and digital photography, lossless compressors cannot achieve the level of compression needed for digital video. The capabilities of lossy compressors do vary and some codecs are better at doing what is expected than others. The flexibility of digital video allows for distribution on DVD, streaming media and vidcasting on the internet, video Ipods, memory sticks and video Ipods and Play Station Portables (PSP). However, each of the devices and technologies require digital videos to be converted to another file format and compressed further to make the handling of video by these devices economically feasible. The only restriction is any video editing should be done before a digital video is lossy compressed to the file formats used by these devices. Codecs specifically those using lossy compression methods of permanently discarding what is considered unnecessary information through a process or combined processes of Perceptual Compression, Spatial Compression, Temporal Compression, and Data Rate Limiting. How well the codec does its job depends on the quality of the codec and typically there is a trade-off in image quality. A quality lossy codecs can achieve 10:1 compression or more without visible degradation of the image. Perceptual compression takes advantage of the human eye being less sensitive to color than it is to luminance and dedicates less bandwidth to color components. To much reliance on perception compression can lead to unwanted color shifts. Spatial Compression takes advantage of redundancy of similarity in a frame by removing unneeded detail from areas of the image having no detail, such as a clear blue sky, or an empty lake surface. Looking for similarities becomes difficult if there is a lot of unwanted background patterns (noise), interference, and in the case of film to video transfer conversion, film grain. Temporal Compression expands on spatial compression by determining if similarity exists in one frame, perhaps it is shared between successive frames and only proceeds to store what differs between the frames. To achieve further compression, some codecs incorporate motion prediction. Rather than simply comparing two successive frames, this technique notices moving objects in a frame and predicts where they will be in the next frame. Then only the difference between the prediction and the actual location need be stored. The temporal compression scheme makes frames dependant on others around it and this results in editing becoming difficult if certainly-so impossible once it has been done. The ability of such compression is hindered by unwanted background patterns, interference, unsteady camera while capturing images, and in the case of film to video transfer conversion, film grain. Data Rate Limiting is a flexible compression as it can be a constant bit rate (CBR) by varying the amount of compression as required to provide a specified data rate or variable bit rate (VBR) which allows the compression to fit between a fixed minimum and a fixed maximum rate. The rate used is typically determined by the device the video is being converted and compressed to play on, although storage space limits and restrictions can also be a determining factor. The three certainties with digital video are: engineering the devices and software to work as expected is more difficult than using the technology; the quality of the devices and software available ranges from being very affordable to very expensive; and, regardless of what when, where or how it gives the typical person the ability to take pretty pictures and make good video. |
Send mail to
webmaster@cytechandprofservices.com with
questions or comments about this web site.
|
