Previous | Table of Contents | Next |
Gilbert Held
The old adage one picture is worth a thousand words reached the attention of network managers and LAN administrators during the mid-1990s due to the proliferation of image-based applications. Through the use of visual databases, fax store and forward systems, and Internet Web pages, the use of images moved from a curiosity to an expectation. Although a picture may be worth a thousand words, its storage and transmission time can represent a thousand fold increase over the spoken word, or perhaps more accurately, a text description of an image. To further exacerbate network bandwidth problems caused by the use of images is the digital camera, whose use is rapidly expanding throughout many organizations. Providing the ability to directly place images into almost any type of document without having to develop film and scan the resulting photograph makes the digital camera directly and immediately responsive to user requirements. Thus, most organizations can expect their level of image utilization to continue to increase.
Although images are now almost indispensable for incorporation into many applications, care must be exercised when doing so. Accepting software defaults, not considering the use of third party products and other oversights, can literally bring down or significantly disrupt the operation of a network. Thus, the purpose of this chapter is to acquaint network managers and LAN administrators with the major characteristics of images acquired through the use of digital cameras and how their effective operation can significantly reduce the effect of the retrieval of stored images on network bandwidth consumption. In accomplishing the goals of this chapter we will examine the use of a specific digital camera for illustrative purposes. However, readers should note that it is the resolution and color depth of images as well as the format used to store such images that govern their data storage and transmission time, topics we will focus our attention upon as we examine the use of one digital camera.
A digital camera uses a charged coupled device (CCD) as a mechanism to capture the intensity of light it is pointed at, resulting in the matrix of CCD elements forming an image. CCD elements can be considered to represent pixels as their number defines the resolution of the camera, and the color depth of each element defines the manner by which the images taken with a camera correspond to the colors the human eye can visualize.
Most digital cameras have two image capture modes standard and fine. The standard mode, which can be considered to represent low resolution, usually results in the storage of an image using every other CCD element value. The most common standard resolution mode used by digital cameras is 320 by 240 pixel elements. The fine mode commonly results in a 640 by 480 resolution, although some relatively recently introduced digital cameras, such as the Kodak DC-120 and Cannon PowerShot, support high resolutions of 1280 by 960 and 832 by 608 pixels respectively. Since the VGA standard is based upon a resolution of 640 by 480 pixels, that resolution provides a good common denominator as it can be viewed by a larger base of PC users than the higher resolutions produced by newly introduced digital cameras that require the use of super VGA monitors to view the enhanced details provided by the higher resolution. Each of ten digital cameras examined by the author of this chapter supported a 24-bit color depth, a technique commonly referred to as True Color, which provides the maximum level of color that a normal human eye can distinguish.
The vertical and horizontal resolution of an image multiplied by its color depth provides the amount of data storage required for an image. For example, a high resolution 640 by 480 image captured using a 24-bit color depth would require 640 × 480 × 24 bits/8 bits/byte, or 921,600 bytes of storage without considering the use of a few additional bytes that specify the format used to store the image.
Since most digital cameras have a limited amount of memory, typically 2, 4 or 8 Mbytes, without a compression method a user would only be able to store a very limited number of images prior to having to delete or download images from their camera. The most common compression method used to store images is the Joint Photography Experts Group (JPEG), a lossy compression method that can significantly reduce the data storage required per image to one-fifth or less than its non-compressed requirements.
In preparing this chapter this author used the Minolta Dimage V digital camera which also uses JPEG compression. For readers not familiar with JPEG compression, it should be noted that this compression method is based upon the use of a Cosine transformation process that reduces blocks of pixels to coefficients that have values of 0 and 1 and compares the contents of those blocks. By specifying a quality factor a user can control the comparison of blocks. That is, on a scale of 0 to 100 a value of 100 results in two blocks of pixels being considered equivalent only when all pixels are identical. In comparison, lower quality values result in blocks being considered to be equivalent even though they differ by a greater number of pixels as the quality value decreases.
Since a Cosine Transformation process is used to reduce the values of pixels in a block to a series of 1s and 0s for comparison purposes, the use of a quality factor of 100 results in certain groups of blocks being considered equivalent even when they actually differ due to slight color depth differences. Thus, the use of a quality value of 100, while providing a minimal amount of pixel loss upon image reconstruction, can result in the storage requirements of an image reduced by a factor of four or more. That is, an image that might otherwise require 1 Mbytes of storage when compressed using JPEG with a quality value of 100 may require less than 250,000 bytes of storage.
The Minolta Dimage V digital camera stores images in its memory using a default and non-alterable JPEG quality value of 100. This means that although images are stored using a JPEG format, that format only represents a good starting place as a slight reduction in the quality scale can result in additional data storage reductions that, while essentially unnoticeable to the human eye, can also reduce the transmission time associated with moving an image locally on a LAN or to a distant user via a wide area network connection. Since the best way to become familiar with the potential effect of a digital camera upon network operations is by its use, lets do so.
Previous | Table of Contents | Next |