Sungkyunkwan University v. LMI Technologies (USA) Inc. (N.D. Cal. 2017)

Method for Processing Images from 3D Camera System Found Invalid under 35 U.S.C. § 101

By Joseph Herndon --

It is well-known law today that under 35 U.S.C. § 101, a patent claim that recites a solution to a problem but not the means of achieving it is likely not drawn to patent-eligible subject matter.  Courts have found that such claims, rather than disclosing a concrete solution, seek to monopolize the very idea of a solution itself.

In the present case, Sungkyunkwan University sued LMI Technologies in the U.S. District Court for the Northern District of California for infringement of its U.S. Patent No. 7,957,639.

The '639 patent is directed to determining an optimal exposure of a structured-light based 3D camera.  The 3D camera of the '639 patent includes a camera and a projecting unit such as a beam projector that illuminates a predetermined pattern on an object, captures an image of the object with the pattern illuminated thereon, and obtains 3D information by analyzing the obtained pattern.  The '639 patent describes that the precision of the calculated 3D data depends on how accurately the pattern is discriminated from the captured image.  To obtain optimal exposure for pattern discrimination, an image captured from a projected pattern without spread phenomenon (saturation) is compared with an image captured from a non-projected reference scene, and a pattern may be optimally discriminated when the intensity difference therebetween becomes the maximum.  As a scale for the intensity difference, SNR and difference in brightness are used.  The '639 patent describes that to determine an optimal exposure level of a camera according to dynamically changing environmental factors, the structured light based camera system employs a method for automatically determining an optimal exposure level by analyzing an image captured at a predetermined exposure level.

Claim 1 is representative and is reproduced below:

1.  A method for determining an optimal exposure of a structured light based three dimensional (3D) camera system including a projecting means for illuminating a predetermined pattern on a target object, an image capturing means for capturing an image of the target object with the pattern projected, and a processing means for reconstructing 3D data based on the captured image, wherein the method comprises the steps of:
    obtaining an image of a target object with a predetermined pattern projected thereon, which is illuminated from a projecting means, and an image of the target object without a predetermined pattern projected thereon; and
    determining automatically an optimal exposure level of the structured light based 3D camera system using said two kinds of images.

LMI filed a motion to dismiss the case based on claim 1 being directed to an abstract idea.  The District Court agreed, and found that claim 1 of the '639 patent is directed to an abstract idea.

To begin, the District Court noted that the claim's preamble recites the technological context of the invention and its end goal; the first element restates the technological context; and the second element restates the end goal.  Following, the District Court found that the method of optimizing exposure on a structured-light 3D camera -- the "how" that's nominally the subject of the claimed invention -- is absent.

The closest claim 1 comes to disclosing how the invention intends to optimize exposure is its reference to two images generated by a structured-light camera (e.g., "determining automatically an optimal exposure level of the structured light based 3D camera system using said two kinds of images").  But as the patent specification makes clear, the images aren't the subject of the invention -- "patterned" images are an inherent part of the prior-art cameras, and "unpatterned" images are just ordinary images.  The District Court found that the claim's repeated reference to the steps involved in using a structured-light camera adds nothing of substance to the method that's actually being claimed here.

By describing the function of the prior-art cameras, the University situated the claimed invention in the technological environment in which it operates, but it did not explain the invention itself.

The overall result is to place the full weight of patent-eligibility on the claim's use of the word "using" (e.g., "determining automatically an optimal exposure level . . . using said two kinds of images").  But whatever method might be encompassed by the idea of "using" patterned and unpatterned images, claim 1 doesn't disclose it.

The District Court found that the only effect of the word "using" is to disclose the starting point of a method that the claim doesn't explain:  "using" two images, you get an optimal exposure, or rather "using" certain inputs, you get a certain result.  The District Court found that this is not enough to make claim 1 concrete because a claim cannot recite inputs and end result and attempt to claim every way of arriving at the end result.

The District Court noted that the patent specification in fact teaches the steps involved in optimizing exposure on a structured-light camera.  But claim 1 does not recite such steps.

Finally, the District Court noted that at oral argument, counsel for the University made vague reference to the fact that the '639 patent is translated from a Korean patent, as if to suggest that some concrete aspect of claim 1 was lost in translation.  But the District Court was quick to note that it's difficult to understand how this helps the University, because it's the U.S. patent that the University is asserting.

Thus, the District Court granted the motion to dismiss.

The end result here is not too surprising given that the claim lacks details of how the method is performed to achieve the recited result.  A claim that recites to obtain two types of images, and then to automatically determine an exposure level using the two kinds of images simply lacks enough detail to satisfy 35 U.S.C. § 101 under today's standards.  Although not discussed by the District Court, the term "automatically" also tends to offer no help in these instances since the claim fails to describe how the function is automatically performed as well.

Sungkyunkwan University v. LMI Technologies (USA) Inc. (N.D. Cal. 2017)
Order Granting Motion to Dismiss by District Judge Vince Chhabria

Comments

[Paul Cole]

In the UK, automaticity was held not to be a basis for patentability in 1909 in British United Shoe v Simon Collier which is a classic case taught to students of patent law: see also the CIPA Guide to the Patents Acts 8th Ed, Sweet & Maxwell, London at 3.27. Brief reference to the MPEP reveals similar holdings in 19th century cases by the US Supreme Court, and the cases cited in the present opinion merely reveal the same principle in a modern context. The present decision is therefore, unfortunately, unsurprising.

Readers of the patent will note that details are to be found in claim 2 which was not in issue.

[EG]

"It is well-known law today that under 35 U.S.C. § 101, a patent claim that recites a solution to a problem but not the means of achieving it is likely not drawn to patent-eligible subject matter."

Hey Joseph,

The issue you describe with such patent claims should be addressed under 35 USC 112 (a)/first paragraph as a lack of "written description" (or possibly "enablement") issue and/or under 35 USC 112 (b)/second paragraph as an "indefiniteness" issue, not 101-that's what O'Reilly v. Morse actually held, contrary to what many others have incorrectly said otherwise. That's also the egregious harm the nonsensical and broken Mayo/Alice framework has done to what was previously a claim construction issue under 35 USC 112.

[Paul Cole]

@ EG

With the utmost respect, the following requirements can be derived from Section 101:

Novelty
Invention or discovery
Utility
Categorical compliance:
* Process
* Machine
* Manufacture
* Composition of matter.

If it is not plausible that an invention has been claimed, then section 101 rejection is available.

In Mayo the novel features were the upper and lower limits of a therapeutic window for thiopurine drugs, which were mere items of information not falling within the "process" category because they were not transformational. The remaining two features lacked novelty, and although raising 102 issues were disqualified from adding to eligibility under 101. As an ordered combination those features added nothing to the first two features since the only acts recited in the claim were old, and the novel steps recited no actions to be taken and took the old test no step further. In the present case the main novel feature of the claim added nothing of an inventive character since it merely recited solving a problem without indicating how that was to be done. Nineteenth century law, as I have pointed out, indicated that this was not sufficient to amount to eligible invention (nowadays 101 not 103).

[EG]

Hey Paul,

With all due respect, in the US you should still start with 35 USC 112-until you understand what is in the claim "as a whole," you can't make any judgment as to what the claimed invention is and whether its really a "process" or not. Also, the fact that the steps recited in the claimed process may be "old" does not mean that 101 is implicated-instead, you should be considering whether the claimed process (also "as a whole") is novel (under 35 USC 102) and if novel, for obviousness (under 35 USC 103)-the fact that SCOTUS has screwed up the normal order of statutes (112, 102, 103, and finally 101) by looking at 35 USC 101 first, using the nonsensical Mayo/Alice framework, and failing even to define what they mean by "abstract idea" is why we're in the current chaotic situation.

[skeptical]

Mr. Cole,

At the risk of (yet again) telling you so, novelty is not a 101 requirement.

[skeptical]

Further, Mr. Cole, you appear to not understand what our Congress did in 1952, or why they did what they did.

Prior to 1952, the Court had the common law power to develop the definition of "invention."

Not so after 1952, as it was at that time that our Congress removed that power and reset the requirement to be non-obviousness. It was at that time that the pre-1952 single paragraph was broken up into separate and distinct sections of law.

You continue to want to view our law under a prism of pre-1952 understanding, and do not give proper weight to what happened in 1952 - and perhaps more importantly, why what happened in 1952, happened. To refresh your memory of the why, I point you to the self-describing (but certainly in an unflattering manner) phrase:

"The only valid patent is one that has not yet appeared before us."

As you are evidently a student of history, it is rather surprising (and ultimately disappointing) that you continue to not grasp ALL of the important history as regards section 101 of our law.

[Paul Cole]

@ Skeptical

As a student of US case law under Section 103 since 1976 and as a concerned campaigner on Section 101 since the Prometheus decision in 2012 I am aware of the intentions of those who drafted the 1952 Act and revere the memory of Pat. Federico and Giles Sutherland Rich.

From the standpoint of the judges, the important consideration is not the legislative intention but the statutory language passed by Congress. If you read the Graham v John Deere decision, you will find statements that in passing section 103, Congress did not intend to change the obviousness standard, and although the flash of genius in Cuno was eclipsed, much earlier case law remains pertinent. The words "invents or discovers" have not vanished from Section 101. Similarly, although there is a novelty section 102, the words "new and useful" are still in Section 101. Anyone trying to argue that the words have no weight from the eligibility standpoint has an uphill battle and much contrary precedent to overcome. The statutory requirements are those listed in my earlier comment, and if Congress intended to remove novelty and obviousness from section 101, the answer must be that their intentions proved only partially successful.

I re-read Mayo with the utmost care in preparation for some comments I offered at a BIO conference a month or so ago, and the analysis I gave so far as I can possibly tell reflects the heart of the decision. There were two non-transformative ineglgible features and two old features in the claim. Much trouble and many wasted words would have been avoided by a holding that the claimed subject matter was not eligible because the novel features straightforwardly did not fall within the "process" category and the remaining features added nothing eligible.

[skeptical]

Mr Cole,

"If you read the Graham v John Deere decision, you will find statements that in passing section 103, Congress did not intend to change the obviousness standard"

That is not quite true.

What I read in Graham is a LOT of dicta by a Supreme Court spinning and attempting to regain a power that was stripped from them.

Do not fall to the "EVERYTHING said by the Supreme Court is Gold" ruse.

As to "The words "invents or discovers" have not vanished from Section 101."

That would be correct - but as I have mentioned, the power of the Supreme Court to change the meaning of those terms through Common Law evolution WAS stripped from the Court. The Court attempting such truly has no force of law.

As to "Similarly, although there is a novelty section 102, the words "new and useful" are still in Section 101."

Again (the echoes of "I told you so" rumble on by), the presence of the words in 101 DO NOT mean that 101 is a controlling section of law on what those words mean. Congress acted with a purpose to separate the prior single paragraph. It is just not within the authority of even a "supreme" Court to set aside what Congress has done (notwithstanding that, the Court will certainly try to do so - but those attempts are [and should be] singled out for the tragedy that they are).

As to " The statutory requirements are those listed in my earlier comment, and if Congress intended to remove novelty and obviousness from section 101, the answer must be that their intentions proved only partially successful."

Agreed. The Court HAS been unusually successful in achieving its desired ends. That success though is NO indication of correctness.

As to "Much trouble and many wasted words would have been avoided by a holding that the claimed subject matter was not eligible because the novel features straightforwardly did not fall within the "process" category and the remaining features added nothing eligible."

Well, I would be more than happy - ecstatic even - if our Courts took as much care and as much reverence for the Rule of Law as you have shown.

Sadly, our Court has other designs.

[Paul Cole]

@ Skeptical

My initial impressions of Prometheus v Mayo are found in postings I made at the time, see

(1)Mayo v Prometheis, a European view http://www.patents4life.com/2012/03/mayo-v-prometheus-%E2%80%93-a-european-view/,

(2) Prometheus v Mayo, a European view, https://patentlyo.com/patent/2012/04/guest-post-prometheus-v-mayo-a-european-view.html and especially

(3) Prometheus v Mayo, The Wrong Rat, http://www.ipwatchdog.com/2012/05/06/prometheus-v-mayo-the-wrong-rat/id=24622/.

The comments on the Wrong Rat paper remain of interest, especially those on Titanic and the respective knowledge of Captain Smith and Thomas Andrews and the utility of that knowledge in terms of the saving of life. On reconsideration I can see a transformative element to the interpretation clauses in the Prometheus claim that could have led a more benevolent Supreme Court to take a different view, since the correct interpretation of the Prometheus test, just like knowledge of the effect of progressive breach of the compartments in Titanic, saves lives.

At the time when I first read Mayo, I had some 40 years of experience in reading case law and ferreting out the principles of law that could be derived from them. My conclusion, expressed in postings at the time, was that no workable principle of law could be derived from its holdings.

Your point on the relationship of 101, 102, 103 and 112 is well taken. However, a more informed view needs an expert on US statute law interpretation, and notwithstanding the excellent Scalia and Garner work which is on my bookshelf, an individual with deeper relevant knowledge (including law outside the IP field) is needed.

On reflection, automaticity in the present case falls into the wrong category, 103 or 112 but not 101. The Wrong Rat principle applies.

[Paul Cole]

For convenience of reference this is what I commented in relation to my Wrong Rat posting about Titanic and which still resonates here:

Consider the situation on-board Titanic immediately after it hit the iceberg. The Ship’s carpenter and Captain Smith tested to see whether the ship was making water. They lowered probes into several compartments and measured the height to which the probes were wetted. But they did not know whether or not the ship would sink. Thomas Andrews was the designer, had done the necessary calculations and may have made models. He knew that if four or less compartments were flooding the ship would continue to float and with five or more compartments flooding it would sink. With that knowledge he was able to tell Captain Smith that the ship would indeed sink quite soon and that lifeboats should be launched. Whatever was the loss of life on Titanic, it would have been greater without that knowledge because launch of lifeboats would have been delayed.

Prior to the patent in issue the state of knowledge was analogous to the ship’s carpenter or Captain Smith – they could do tests but did not know how to interpret them. With the further information in the patent the level of knowledge had advanced to that of Thomas Andrews – the analogy is indeed close because the risk of incorrect administration of thiopurine drugs in at least some patients is a risk to life.

In reply to Muenchen’s post, the potentially novel feature of the final version of the claim is the hitherto undiscovered link between an immune-mediated gastrointestinal disorder, a therapeutically effective dose of a drug providing 6-thioguanine, risk of leucopenia and a level of 6-thioguanine within the range 230 – 400 pmol per 8×10^8 red blood cells. That kind of link has commonly been recognized in the EPO as amounting to a technical feature (e.g. by the Enlarged Appeal Board in Mobil Oil III/Friction reducing additive), and is also recognized by the USPTO, I believe, in medical treatment claims. It would be a serious consequence if this kind of link was held to be insufficient to support patentability, but it is difficult to see on the principles set out by Justice Breyer where the ambit of s. 101 ends. Leaving it to arbitrary decisions (the length of the Lord Chancellor’s foot) is not helpful.

[skeptical]

Mr. Cole,

The reference book sitting on your bookshelf aside

"(Your point on the relationship of 101, 102, 103 and 112 is well taken. However, a more informed view needs an expert on US statute law interpretation, and notwithstanding the excellent Scalia and Garner work which is on my bookshelf, an individual with deeper relevant knowledge (including law outside the IP field) is needed.)"

what exactly is the deeper relevant knowledge that you think is needed, and why is such needed?

If you are going to disagree with my point (after praising it), I would like to know why you are disagreeing with it.

[Paul Cole]

@ Skeptical

The ABA has recently written to Michelle Lee reiterating the view that you would take that novelty is not a 101 concern because it is now covered by 102.

I don't either agree or disagree with you on your point. I simply don't know. I am not sure whether even the Supreme Court justices and their law clerks properly know. We need the views of a top flight academic or other legal expert who specializes in statutory interpretation.

[EG]

"What I read in Graham is a LOT of dicta by a Supreme Court spinning and attempting to regain a power that was stripped from them."

Hey Paul,

I completely agree with Skeptical in this statement. Indeed, if you haven't done so, I would suggest you read Adam Mossoff's opus entitled WHO CARES WHAT THOMAS JEFFERSON THOUGHT ABOUT PATENTS REEVALUATING THE PATENT “PRIVILEGE” IN HISTORICAL CONTEXT: it's eye-opening about how poorly SCOTUS understands even the basic history surrounding the Patent and Copyright Clause in the U.S. Constitution at best, and at worst, is grossly distorting that history to support the dicta in Graham that Skeptical noted. And all or most of that dicta is frankly not supported at all by the actual history surrounding the creation of the Patent and Copyright Clause as Adam (a far more able historian than any of the SCOTUS Justices) so amply demonstrates in his article. In fact, SCOTUS has many, many problematical "skeletons" in its Constitutional case law "closet," several of which are very dark indeed.

[skeptical]

EG,

We would be F A R better off if Congress were to exercise its Constitutional power of jurisdiction stripping and strip the Supreme Court of the non-original jurisdiction of patent appeals.

Couple that with the creation of a new (and UNTAINTED) special Article III patent court (to preserve the actual holding of Marbury), and we could be rid of the nose of wax mashing...

Of course, that would require Congress to not only care, but to recognize both what is within their power and WHY they might want to use that power.

Of that tall order, sadly, I remain....