Digital Photography on Microscopes
Digital photography is advancement in the investigatory field to study images of objects under the microscope. Digital photography is simply the replacement of a film type camera with an electronic camera such as a video camera. The key to digital photography is that no film is involved. The image is transferred directly from the camera to either a digital storage medium, such as a disc, or to a computer for some type of manipulation prior to storage on the computers storage disc. While these definitions make clear that there are very different procedures involved, the key that ties all together is the computer. Most current product introductions to anatomic pathology actually are combinations of digital photography image capture and image analysis. Fully automated microscope systems that require little or no human intervention are still a dream of some, and a nightmare to others.
Electronic cameras, that is, cameras that do not use film, are often called video cameras. This is not an accurate term, since many electronic cameras do not capture images fast enough to provide a video signal. All modern electronic cameras are based on semiconductor type detectors, in contrast to the older vacuum tube technology, as exemplified by the vidicon tube. Modern cameras primarily use semiconductor detectors of a type called CCD although a new style called CMOS or Complementary Metal Oxide Semiconductor is rapidly being refined. CCD detectors have been available longer, and are therefore more highly refined than CMOS detectors, which are a newer technology. At the current state of development, CCD detectors have lower noise than CMOS detectors.
CMOS detectors are generally cheaper to manufacture, and offer the potential for reduced support electronics, making CMOS cameras cheaper to produce. From the imaging point of view, the difference in these two types of detectors is not important for our discussion, since each can provide an image. The essence of all light detectors is a target or area on which light falls. This target must respond to different amount of light by generating a signal. The generated signal should vary depending on how bright the light is. This is exactly how the light meter works in a camera, that is, the needle deflect farther the brighter the light. A light meter cannot capture an image because there is only one light detector. In an electronic camera, the detector target is made up of many small detectors, arranged side by side in rows and columns. A lens projects an image on these detectors, and each produces a signal that is directly proportional to the amount of light falling on that detector. By reading out each individual detector, and then displaying each detectors value, in the same arrangement as the detectors themselves, an image is seen on the display device. Note that this image is strictly light and dark or black and white or monochrome. The individual detectors or sensors of an electronic camera are called pixels, which stands for picture elements. The individual pixels that make up an image can be readily seen on a computer screen if an image is magnified or zoomed sufficiently. Eventually, the image detail will disappear and the image will take on the appearance of a tiled mosaic.
As impressive as this number of pixels is, it is not very many when compared with the information in a microscope field of view. Detail seen in a microscope depends on the quality of the lenses used, and the color of the light used for illumination. Under ideal conditions, very high quality microscopes can resolve individual points in an image that are 0.2 microns apart. This is for a high quality oil immersion lens. A laboratory grade microscope with a 10x objective should routinely resolve image points that are on the order of 1 micron separation. When you think about how many microns one can see in a low power microscope field, it is obvious that dividing this large number by 640 pixels will mean a significant loss of detail. Keep in mind that an individual pixel sensor only provides a single number based on the average amount of light that falls on it, regardless of the amount of detail that makes up that light. The individual pixel sensors that make up a camera detector can have a square shape, or a rectangular shape. The rectangular shape is commonly used in home video cameras, and many broadcast television systems. For these uses, the shape of the sensor is not critical, since the eye does not readily perceive that the resolution in the short direction of the sensor is greater than in the long direction. For microscope use, it is preferable to use sensors with square pixels. This assures identical resolution in both the x and y directions, and also makes it much easier to calibrate the camera and microscope for specific types of image analysis tasks.
