Background
Digitech Image Technologies held U.S. Patent No. 6,128,415, which covered a “device profile” for use in digital image processing systems. The device profile was essentially a set of two data sets: one describing the color characteristics of a device (such as a scanner or printer) and one describing its spatial characteristics. The idea was that if you knew the color and spatial profile of each device in a digital imaging chain, you could transform images accurately between devices — ensuring that colors scanned by one device printed accurately on another. This “device profiling” concept was central to color management software.
Digitech sued dozens of defendants — camera manufacturers, printer companies, software vendors — for infringement. The Central District of California consolidated the cases and granted motions for summary judgment of invalidity under § 101, finding the claims covered ineligible subject matter. Digitech appealed.
The Court’s Holding
Judge Reyna, writing for the Federal Circuit, affirmed invalidity. The court addressed two types of claims: device profile claims and method claims.
For the device profile claims, the court held that a device profile — a collection of numerical data sets describing device color and spatial properties — was not a patentable “thing” under § 101. The categories of eligible subject matter are processes, machines, manufactures, and compositions of matter. A pure collection of information — numbers and data relationships — falls into none of them. The device profile existed only as abstract information; it had no physical embodiment. Patent law does not protect “information” as such, the court held, regardless of how specific or useful that information is.
For the method claims, which described steps for generating the device profile using mathematical formulas, the court held the claims were directed to mathematical relationships and algorithms — abstract ideas under § 101. The claimed methods described a process of organizing information through mathematical correlations without any tie to a specific machine or a physical transformation of matter.
Key Takeaways
- Pure data structures — collections of information with no physical embodiment — are not patentable under § 101, regardless of their commercial value or technical utility.
- Methods that consist primarily of mathematical calculations or correlations, without tying those calculations to a specific machine or producing a physical transformation, are abstract ideas ineligible for patent protection.
- The decision is a direct application of the principle that mathematical relationships themselves are not patentable — a principle dating back to Diamond v. Diehr (1981) and Gottschalk v. Benson (1972).
- Image processing and digital photography companies relying on broadly drafted data-structure patents received a clear signal that their patents face § 101 vulnerability.
Why It Matters
Digitech contributed to the post-Alice clarification that the § 101 prohibition on abstract ideas extends to a broad range of digital data concepts — not just financial software or business methods, but any patent that essentially claims a set of numbers, correlations, or information structures without a concrete physical anchor. The decision was particularly significant for the digital imaging, photography, and graphics technology industries, where patents on data formats and calibration schemes were common.
More broadly, Digitech reinforced a principle that patent prosecutors need to keep in mind: claims to data or information must always be tied to a specific hardware implementation, a particular technological use, or a physical transformation to have a chance at surviving § 101. Claiming the data itself, or mathematical relationships among data, is a path to invalidity.