A no-noise approach to color imaging
Technologies attempting to improve digital color imaging have mainly focused on increasing sensor resolution by adding more megapixels. Now, however, NuCORE Technology (Sunnyvale, CA; www.nucoretech.com) has introduced its CleanCapture technology that emphasizes the elimination of noise by two novel image processors that boost image color quality. This technology integrates a preprocessor and a postprocessor to dramatically minimize noise and digital artifacts during image processing. The result is that digital still and digital video camera sensors can capture truer representations of the subject being imaged.
The company's patented NuCORE Dynamic range eXpansion (NDX) 1260 digital image preprocessor chip provides analog signal conditioning and data preprocessing for CCD sensors to ensure that the follow-on Smart imager Processor (SiP) 1270 digital image postprocessor chip receives high-quality pixel data without digital artifacts. The NDX-1260 integrates a dual-differential correlated double sampler; pixel-by-pixel programmable gain amplifier; high-speed dual-pipeline analog-to-digital converter with a programmable output data bus; and advanced black-level autocalibration circuitry. The back-end SiP-1270 processor chip uses a flexible hard-wired image-processing pipeline to embed imaging algorithms and deliver a 75-Mpixels/s throughput. It also includes a high-quality-motion JPEG engine that can support up to 24-Mpixel CCDs, high-speed compact flash and card memory interfaces, digital zoom, USB 1.1 interfaces for hard disk and minidrive storage, and an embedded 32-bit ARM-9 microprocessor.
Inherent noises, which include cross-color, false-color, color-phase, random, fixed-pattern, line, and digital artifacts, can degrade image quality. Many video-system designs use digital-processing techniques to minimize the visual distractions caused by noise. This approach is almost always a trade-off between the effect and the desired image resolution or sharpness. NuCORE's approach is to eliminate all sources of noise with special circuits so that neither the camera sensor nor the user needs to minimize the visual distractions caused by noise sources after the image has been acquired.
Image-processing benefits include artifact-free film-quality images with wider dynamic range for still image and video capture; analog white balance for greater resolution in shadows and bright areas; no digital artifacts and false colors caused by digital white balance; clean, smooth tonal and artifact-free images; minimal shutter delay and no line noise; support for current and next-generation CCD sensors; and markedly reduced file sizes for digital image data storage and data transmission.
Based on an API software architecture, OEMs can program a user interface and an on-screen display preference without direct knowledge of the underlying image-processing hardware. A source code license is available. Using these processors OEMs can get their color imaging products to market quickly and yet add their own value.
As CCD sensors steadily increase to ever-higher megapixels of resolution, the associated image-processing subsystem is emerging as the critical stage of system performance when high-quality color is a major requirement.
George Kotelly
Editor in Chief
[email protected]