Dust and Electrostatic Charges Removal

Services included in our On-Line offer
Our basic service includes the dust removal of film by ventilation of ionized air at low pressure. By eliminating the electrostatic charges on both sides of the films, this non-aggressive process allows to evacuate almost all fibers and dust. It is efficient enough for film strips that have been properly preserved.

Services not included
However, we can not ensure the eradication of scratches and dust stuck on films. The handling or preservation defects of the films usually involve additional digital restoration works that can be quite time-consuming. The restoration works, to achieve at post-processing time, can significantly impact the unit cost of a complete scanning service. In addition, image editing software tools dedicated to this purpose are easily accessible to the public.

We have chosen not to unnecessarily increase the basic prices of our services. This intervention of manual retouching in post-processing, technically accessible by the majority of our customers, is not integrated in our standard On-Line offers. It can, however, be the subject of an option on request.

Exceptional situations
Stained films, usually by fingerprints, may require cleaning prior to scanning. This operation presents a risk of deterioration of the silver emulsion. We also do not provide this kind of service as part of our On-Line offers.

Notes about the implementation of the Digital-ICE technology

Unlike low-cost mass-market scanners or modular scanning heads that can be integrated into mini-labs systems, Hasselblad has made the choice to not integrate the Digital-ICE process in its high-end Flextight scanners.

Here is a summary explaining the theory behind Digital-ICE and why the implementation of this technology on devices bearing the "Digital-ICE" symbol remains approximate.

Usable only on certain types of film, the Digital-ICE process can be relatively effective for small dust. In return, it is not free of side effects. In particular, the sharpness of the final image is indeed diminished. If this degradation seems relatively invisible on scans made at low or medium resolution, it is no longer the case on scanned images with the resolution and optical sharpness of the Flextight X5.

The theoretical principle of the Digital-ICE process is to try to create a binary image that is supposed to delimit the areas of the image affected by the presence of unwanted residual particles and scratches present on the film, then to exploit this binary image as a "defect mask".
To achieve this, an infrared source is inserted into the illumination block, an additional sequence for IR flashing is inserted in the scanning process, the scanner is then supposed to reconstitute an anti-scratch mask on the basis of the IR image obtained. Given that dust, like scratches, reside by nature on both sides of the film, they occupy their own focusing planes, which are obviously different from that of the useful image (gelatin layer). Moreover, in order for the digital-ICE process to be technically viable, the optical block design should not introduce any differential geometric distortion between the two captured images, i.e. between the IR image and the wide spectrum RGB image. Neither the optical acuity nor the depth of field should depend on the ROI (Region Of Interest) in the image. The only way to do this would be to impose telecentricity, so that the main rays can be perpendicular to the object plane over the entire surface of the image to be scanned on the film. It is at this price that one could really hope to obtain a perfect coincidence between the IR defect image and the raw RGB image. But, at comparable optical resolution and sharpness, telecentric optics are much more expensive and voluminous than specialized macro optics. On the other hand, the choice of telecentricity would restrict the scanner to a fixed image size determined by the optics. It is indeed not possible to vary the magnification ratio of a telecentric optics. To support different formats of images to be scanned, this would also involve integrating a mechatronic device to carry several telecentric optics with different magnification ratios. Note also that the scattering of the illumination beam should be different for IR and for RGB illumination...

Unfortunately, this is not the kind of implementation found in scanners with digital-ICE technology. These devices thus come up against the extreme difficulty of reconstructing an anti-scratch image which is sufficiently sharp, undistorted and coinciding with the raw RGB image.

Another difficulty comes from the supposed transparency to the IR of films with silver emulsion. This problem results in a more or less significant degradation of the signal-to-noise ratio of the IR image, thus leading to false numerical interpretations of scratches and dust.

Finally, the technical limitations due to the implementation of this technology by the scanners bearing the digital-ICE acronym always lead to compensate for the shortcomings by a slight blurring of the resulting image. In addition, sporadic erasures of useful details are also noted.

Only the old Nikon Coolscan-9000-ED scanner (obsolete) implemented this technology pretty much correctly. However, even with this device, the loss of sharpness was compensated for by a contour correction (digital sharpening), making the final image relatively hard when the digital resolution was set to the maximum possible offered by this device. To finish with the Digital-ICE, add that this process does not work with black-and-white negatives and at best very bad with Kodachrome reversible.