Early attempts to develop video cameras with resolution and sensitivity as good as the human eye--in theory, the ultimate aim--were not very successful. This is because the eye presents a three-dimensional image to the brain, which uses a high degree of parallelism in image processing. The brain's ability to interpret the image contents at the same time also contributes to very sophisticated and intelligent image optimization.
The application of digital technology within the camera has come about mainly from the need to improve image quality, sensitivity, and dynamic range. DSP technology and advanced CCD chipsets produce an exceptionally high number of gray levels, resulting in wide dynamic range and high-quality image reproduction of details in high- and low-lit scenes. Clearly, the camera's imaging performance, including imaging and control flexibility, are critical when it comes to making a choice.
The quality of a CCD camera image is directly related to the number of gray levels that can be processed, which is governed by the DSP's architecture. For example, a 1-bit processor can only deliver an image similar to that produced by a matrix printer, while a 4-bit processor delivers 16 gray levels to produce a somewhat better image (see the images below).
Most security cameras use 10-bit processors and are potentially capable of producing pictures with 1024 gray levels, giving relatively smooth variations from light to dark. Bosch DinionXF cameras employ 15-bit digital signal processing, which delivers a 32-fold increase in the number of gray levels and color accuracy compared with a 10-bit DSP, producing what the eye interprets as a virtually continuous variation of gray level. This results in exceptionally accurate picture reproduction with broad dynamic range and vivid colors over a broad spectrum.
The DinionXF camera's 15-bit digital signal processing gives a virtually continuous variation in gray level from light to dark
