Raw Forge is a tool for photographers whose cameras support CCD-native raw images. Most major leading brands will be supported, but support for recent Fuji models (like the S3) might take longer.

This will be of particular interest to amateurs* wanting to try astrophotography, who have been disappointed by the output offered by their camera. It is an increasing trend for lower-quality, higher-resolution CCDs to be used within cameras. This means that images with long exposure (like night sky shots) will show background noise and unwanted bright spots. Raw Forge will take your CCD images, eliminate most of the noise, and present a surprising level of detail that you would not have seen in the original image.

The project is in the early stages of development, and a release looks likely at this stage. For this reason, details here are concerned only with basic specification and potential applications.

Compatibility

Raw Forge currently works with Nikon Raw (NEF) images. It is planned to adopt Adobe DNG format next, and approach other proprietory formats if the need is identified: RAF, RAW, CRW, CR2, DCR, MRW, ORF, PEF, SRF.

Noise Reduction

Raw Forge will feature noise reduction that hides the CCD faults and environmental noise associated with CCD (digital) photography. At present, most digital cameras do not clean up faulty CCDs. A raw-resolution noise reduction method has been tested for Raw Forge, which is superior to the 'noise reduction' or 'clear image' features found in some digital cameras.

Image adjustments

Basic image adjustments will be possible. The first release will feature CCD-native level adjustment. Further releases will allow adjustment in many colour spaces (CCD-native, RGB, HLS to begin with), and allow conservation of saturation.

Image editing

Further releases will include painting tools, image filters, and layer-based editing. It is known that many image editors support these features, but Raw Forge will support these at CCD-native resolution, as well as RGB and other colour spaces.

Compositing

Three types of compositing have been identified: master-detail large format images, adjacent stitching, and stacking. Stacking will be implemented first, for astrophotography applications, but the first release of this feature will not automatically align the stacked images.

This screenshot shows the results of a simple filter to remove 'Hotspot' CCD defects, using a black-point picture (shown top-left), and applying it to the source image (bottom-left), to obtain a clean image (bottom-right).

Although you have to look carefully to see the difference here, the difference is most apparent in images converted from Raw, e.g. by the Nikon View suite, where errors are spread further.