Digital still cameras could be dubbed the Polaroid of the 90s. They provide quality electronic images in a jiffy for Web pages and word-processing documents. Many consumer digital cameras are breaking the million-pixel mark, meaning they can produce detailed, aesthetically sized 5 3 7 images, one of the most commonly ordered reprint sizes, according to Eastman Kodak Co., Rochester, N.Y. With recent models, Konica USA Inc., Englewood Cliffs, N.J., Kodak, and Ricoh Consumer Products Group, Sparks, Nev., join the megapixel club. The cameras capture more detail and richer color than their VGA counterparts. More pixels, referred to as higher resolution, means bigger, better images. With production costs dropping, many high-resolution digital cameras go for less than $1,000.
And they’re all about easy access. Digital cameras have on-camera photo manipulation options and also provide convenient camera-to-computer image transfer. Many come bundled with editing software such as PhotoStudio from ArcSoft, Adobe PhotoDeluxe, and Photovista.
Snappy options abound
Some score with high resolution, but other cameras win points with specialized features making them easy to use. A 180° rotating zoom lens allows photographers to capture varied composition angles on Ricoh’s 1.3 million-pixel RDC-4300. With another Ricoh product, the RDC-300Z, VGA resolution goes trim. The camera measures 4.75 3 1.4 3 2.9 in. — pocket size for convenience. It goes small in another way too. Macrofocus technology lets the camera capture subjects within a focal length of 0.5 in.
Kodak’s DC210 presents its user interface as “intuitive.” Users can turn a dial and keep a finger on the shutter button while going from capture to review, to connect, and to preference settings. Retrieving images is intended to be simple too. When the camera plugs into a computer, it acts as an external hard drive and images can be accessed on the camera just as they would be pulled from a folder on the computer’s hard disk.
Toshiba Imaging Systems, Irvine, Calif., also offers an easy image-access feature with its PDR-2 digital camera. A built-in PC card flips open letting the camera plug into the side of a notebook computer for image retrieval. On the other hand, speed is Konica’s forte. Four images can be shot in rapid succession with the Q-M100. The speed-shot mode was added for play-by-play action sequences.
Let’s get digital
It looks like a “normal” camera on the outside, but what goes on inside a digital camera? Light enters the camera through a shutter, but instead of film preserving the moment, a sensor picks up the light waves. Most digital still cameras on the market contain a charge-coupled device (CCD), a sensor that converts photons to electrons.
The CCD processes the image in analog form, meaning the output signal varies continuously, as contrasted to digital signals that switch on and off. The light is divided into red, green, and blue by a single CCD. This is different from television cameras which have three imaging sensors, one for each color. The sensor contains an array of pixels which captures light and effectively determines the image size, or resolution. Most CCDs are capable of recording up to 16.7 million colors per dot, or per pixel.
Some cameras, such as Kodak’s DC120, use rectangular pixels, but computer monitors and printers read in square pixels, as do most digital cameras. For a sensor that uses rectangular pixels, the computer or printer reads two rectangular pixels as three square pixels. Thus the number of pixels on the sensor may be only 800,000 but the image reads as 1.2 million pixels.
The CCD output drives an analog-to-digital converter which digitizes the recorded light waves into bits of information coded as ones and zeros. The digitized data is then sent to a temporary storage area in the camera. At this point, the red, blue, and green separations are superimposed. Then the image is sent to an accessible memory storage unit, often in a compressed format such as JPEG. Storage units vary from camera to camera, but they’re often removable cartridges, such as 2 or 4-Mbyte memory cards. Many cameras offer a serial connection from the camera to a computer, but this usually requires more time to upload the images than using a removable cartridge.
An energy-saving alternative to the CCD was first used in digital still cameras about a year ago with Newbury Park, Calif.-based Vivitar Corp.’s V3000 and V3100. Complementary Metal Oxide Semiconductor (CMOS) advanced pixel sensors can be produced at a lower cost than CCDs and offer easier integration, lower power consumption, and easier camera design, according to Tom Piehn, principal technologist with Vivitar. CMOS imagers and CCDs function similarly, but with a CMOS imager each individual pixel connects to its own amplification and readout structure which can be randomly accessed. A CCD, comparatively, has a linear readout array, which requires reading all the pixels preceding a region before retrieving information for a specific region. The CMOS imager also offers reduced fixed pattern noise with better image quality, Piehn says.
CMOS design uses the same manufacturing platform as most microprocessors and memory chips, while CCDs use a specialized process. The imager is more cost effective to produce than a CCD since there are 100 to 200 CMOS foundries worldwide, as opposed to about 20 CCD foundries. And only about five of those will process CCDs for other companies, according to Piehn.
A CCD needs support chips for noise reduction, data formatting, and so on, which adds to power consumption. A CMOS imager includes everything, even analog-to-digital converters, on one chip, and uses 30% less power than a CCD. Also, as you move to a larger image size with CMOS, the power increase is lower than with a CCD, claims Piehn.
CMOS imagers, in turn, allow for smaller, lighter products such as the Toshiba PDR, which is about the size of a bar of soap. It boasts VGA (640 3 480 pixel) resolution for less than $500 and runs on 2.4 W provided by a lithium battery. Some CCD cameras require four or more AA batteries.
Light in bloom
For a digital imager, being in full bloom isn’t too rosy. The term blooming refers to what happens when extremely bright light hits the pixel array. The image shows a hazy edge and lack of detail around a bright light source, such as a window or light fixture. Paul Gallagher of VLSI Vision Ltd., Vivitar’s CMOS-imager provider, explains blooming by comparing the pixel array to a room full of buckets. A sprinkler system sprays water (analogous to light), and depending on how the sprinkler system is set up, more water will fall in some buckets than others. Eventually, buckets will overflow and spill into neighboring buckets or onto the floor. The same thing happens when very bright light saturates some pixels — the excess light energy dumps into bordering pixels.
The fact that the pixels spill excess energy can’t be changed, but where it spills can be controlled. Energy drains integrated in the pixel array at the top of each column of pixels, or directly underneath each pixel help prevent blooming. VLSI manages to control blooming enough that when its imager is faced with a fluorescent light fixture, the light’s tubes are visible in the image.n