Google Scores Win as Federal Circuit Nixes Digital Imaging Patent Claims
Longitude’s lawsuit over image correction software fails under Section 101
Google walked away with a clean victory last week after the Federal Circuit upheld the dismissal of a patent infringement suit brought by Longitude Licensing Ltd. The ruling, handed down April 30, 2025, reinforces the courts’ continued skepticism toward patent claims that lean on abstract ideas without providing concrete technical details.
Longitude accused Google of violating four of its patents related to digital image correction. These patents centered around identifying a photo’s “main object” — for example, a person’s face — and applying automated corrections such as color balance, brightness, or sharpness just to that portion of the image. While the claims described steps such as “acquiring correction conditions” and “adjusting picture quality,” the court found that these were vague concepts implemented through generic computer functions.
The Court’s Take: Too Much Talk, Not Enough Tech
Judge Timothy Dyk, writing for the panel, agreed with the lower court’s assessment that the patents simply recited a process of adjusting image quality using computers. The court emphasized that while the concept might sound useful, the way it was described lacked substance.
In short, it was all about what the patent was trying to do, not how it actually did it.
The decision compared Longitude’s claims to others the Federal Circuit has rejected in past cases. For example, in Hawk Technology Systems, the court found similar abstract data processing language — like “determining,” “acquiring,” and “adjusting” — to be too broad. Here, Longitude’s patents suffered the same fate.
Why the Court Didn’t Buy the “Innovation” Argument
Longitude’s lawyers tried to argue that their image correction process was a technological improvement over past methods, especially since earlier techniques often adjusted the entire photo without isolating the main subject.
But the judges didn’t see it that way. They pointed out that humans have long been able to make those distinctions using photo editing software. Putting that same idea into a computer doesn’t automatically make it worthy of a patent. Just using “new data” or applying an existing concept in a software setting doesn’t cut it.
Examples from the Ruling
Claim 32 of one patent described a method where a computer identifies the subject of an image, figures out its visual characteristics, retrieves correction settings based on those traits, and then adjusts the image. But it never spelled out how that process is technically achieved — just that it happens.
Another claim focused on breaking an image into segments and using pixel positions to find the subject, yet still failed to explain any real implementation steps. Again, too much function, too little form.
What This Means for Future Patent Claims
This case follows a growing line of decisions applying the Supreme Court’s Alice framework, which blocks patents that are built on abstract ideas unless they show a specific inventive concept. Here, the Federal Circuit found no such concept in any of the 66 claims asserted.
Even when Longitude tried to argue that the use of correction parameters like shadows, brightness, or sharpness counted as an “inventive concept,” the court remained unmoved. Those are standard photo editing features, not innovations.
All Claims Down, Appeal Denied
Longitude appealed after the district court tossed the case last October. That ruling had found the patents “functional and ends-oriented,” a criticism the appellate court echoed. Ultimately, the judges said the claims failed both parts of the Alice test: they were directed at an abstract idea, and they didn’t offer anything inventive beyond that.
The Federal Circuit didn’t need to review every one of the 66 claims in detail. They found all of them shared the same shortcomings — a reliance on generic data processing and a lack of explanation for how the results were actually produced.
Patent holders hoping to challenge tech giants should take note: courts are cracking down on patents that dress up familiar processes in high-tech language. As this case shows, if your claim reads like a set of vague instructions that could apply to any software, don’t expect the courts to be sympathetic.
The decision is another reminder that it’s not enough to describe what a program does. You have to show how it works — and do it with precision.
