Know How

An image sensor can be regarded as an “array of tiny solar cells” converting photons to electrons. The reason for this conversion lies in the photoelectric effect. Such a tiny solar cell is commonly referred to as pixel. A way to think about sensors used in digital cameras is to think of a 2-dimensional array of such pixels of which each transforms the light into electrons. Later this information is read out, digitized and an image is compiled.

Both CCD and CMOS image sensors start at the same point where they convert light into electrons in form of accumulated charge. The next step is to read out this charge of each cell and to digitize this information to make it computer readable. In a CCD device, thecharge is transported across each cell to the “corner of the chip” and read out serial. In most CMOS devices, there are several transistors at each pixel that amplify and move the charge in a more direct way and/or parallel to the digitizer. The CMOS approach is more flexible because each pixel can be read out individually. Most of today’s high speed cameras are using CMOS sensors.

The resolution is commonly measured in x * y pixels and represents the spatial resolution of a camera. The more real pixels a camera has, the more detail it can capture. Some cameras using spatial interpolation techniques. In such cameras a physically lower resolution is blown up by interpolation. Common true pixel resolution in High Speed Cameras is 1Megapixel (1000 * 1000 pixel)

In fact, each pixel accumulates the total intensity of the light that strikes its surface. In order to get a full color image, the sensors use filters (here a Bayer Pattern) to separate the 3 required colors (the so called primary colors RGB = red-green-blue). Once all three colors have been recorded, they can be added together to create the image displayed on your monitor

It is imminent to control the amount of light that reaches the sensor surface. Too much light result in “white images”, not enough light in “dark images” The amount of light reaching the sensor surface is controlled by the aperture. The aperture is the size of the opening in the lens in front of the camera. A smaller aperture gives you a larger depth of field (the image seems in focus over a longer distance) while a larger aperture gives you a reduced depth of field.

The focal length is defined as the distance between the lens and the surface of the sensor. For C-Mount this distance is set to 17.54mm. Further, the focal length is the critical information determining how much magnification you get when you look through your camera. Increasing the focal length results in a greater magnification (“objects appear closer”) There is a wide variety of lenses available ranging from fixed focus, Zoom lenses to macro lens systems.

The shutter speed defines how much light is allowed to pass through the aperture. Think of a shutter as a window shade. Then, for a defined periode of time the shade (shutter) opens and then closes. The time the shade is open is called the shutter speed. A shorter shutter speed will “freeze your motion” on the scene while taking the same scene with a longer shutter time results in smeary images.

Fps is the short form for “frames per second” and represents the number of images taken by a digital or analog camera. This number is used in conjunction with movie and/or high speed cameras and represents “how many images per second are taken”. E.g. 30fps stands for 30 images taken per second, 1000 fps of a high speed camera represents 1000 images taken per second.

In single sensor cameras where the color is reconstructed by the RGB values of the single pixel the color is calculated by the 3x 8bits of each pixel and results in 24Bit color.

The raw data file are “binary” files storing the pixel information of high speed cameras in digital numbers not performing any color interpolation. This result in the fact that a frame recorded with a high speed video camera in raw data represents the size of the sensor in pixels multiplied by the number of frames recorded. (e.g. with a sensor of 1280 x 1024 pixel recorded at 500 frames / sec for 2 sec results in 1.3 GB of raw data) Playing back a raw data file of a high speed camera by means of software, the color reconstruction of the RGB Bayer pattern is performed “as it plays back” On the other hand we do have AVI files. AVI files contain the information about the reconstructed color. This means for all pixels there are 3 values stored representing the red-green-blue information (RGB) of the pixel.. This results in the fact, that uncompressed AVI files are three times larger than an equivalent raw data file of a high speed camera.

To learn more about a circular buffer and how this will effect your System, please click here: LINK