Image Signal Processor Patents: Leaders, Trends & White Space 2026
- Filings have plateaued, not accelerated. Activity peaked at 26 families in 2024 after climbing from 9 in 2017 — a decade of steady claiming, not a recent land rush.
- Demosaicing and denoising sit on old, heavily-cited art. The most-cited record in this set carries 226 citations and dates back to the local tone mapping wave — new filers are building on top of, not around, it.
- Momentum has gone quiet at the top. Recent-year filings from the largest historical assignees show flat or negative year-over-year movement, opening room for new entrants to claim ground.
Filing growth compares 2021 (22 records) with 2024 (26) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 235 records in scope (CR5), not by the ranked leaders only.
What this landscape covers
This dataset tracks 235 patent families filed against image signal processor pipelines — the hardware and software stages that turn raw sensor data into a usable image, including HDR fusion, demosaicing, noise reduction, and on-sensor processing. Coverage runs from 2015 through mid-2026, with the search anchored to H04N23/80, H04N25/00, and G06T5 so the set stays focused on pipeline-stage claims rather than general imaging or display work.
Filing activity is concentrated in the United States receiving office, with Europe, the PCT route, and a smaller tail across India, South Korea, and China. Publication lags filing by roughly 18 months, so the 2026 figure in any trend line understates real activity still working through the pipeline.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
Filing trends and technology composition
Two views of the same 235 families: how filing volume has moved year over year, and which IPC subclasses carry the claim weight.
A decade of steady, now flattening, filing
Filings rose from 9 in 2017 to a peak of 26 in 2024, with 2022 sitting at the midpoint of 18. The shape is a slow build followed by a plateau rather than a recent surge, and the partial 2026 count should be read with the 18-month publication lag in mind.
H04N and G06T dominate; recognition and AI classes are secondary
H04N (173 records) and G06T (159) between them cover most of the corpus, confirming this is fundamentally a pictorial-communication and image-processing field. G06V, G06K, and G06N appear at meaningfully lower counts — recognition and AI-model claims are present but sit alongside the core pipeline work rather than replacing it.
Shares are the percentage of the 235 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Image Signal Processor Technology Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about image signal processor technology landscape and every answer comes back with the patent numbers behind it.
Try EurekaThe records other filings build on
US9787922B2 — Pixel defect preprocessing in an image signal processor
An image signal processor may include a sensor interface that includes a pixel defect preprocessing (PDP) component that performs an initial adjustment of pixel values for patterned defect pixels in raw pixel data captured by an image sensor. To adjust a patterned defect pixel, the PDP component may apply an interpolation technique to values in a gain lookup table according to the pixel's location in the image frame to determine the gain value for the pixel, and then apply the gain value to the pixel. The PDP component may provide the raw pixel data with the adjusted patterned defect pixels to two or more other modules for additional processing.Filed by Apple, published 2017-10-10 — sits early in the corpus's timeline, ahead of the peak filing years.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20130322753A1 | Systems and methods for local tone mapping | 226 |
| 2 | US20130321679A1 | Systems and methods for highlight recovery in an image signal processor | 51 |
| 3 | US20220164926A1 | Method and device for joint denoising and demosaicing using neural network | 43 |
| 4 | US8830367B1 | Frame manipulation to reduce rolling shutter artifacts | 33 |
| 5 | US9741099B2 | Systems and methods for local tone mapping | 27 |
| 6 | US20150092076A1 | Image Capture Accelerator | 27 |
| 7 | US20150348246A1 | Systems and methods for local tone mapping | 22 |
| 8 | US20170163951A1 | Imaging apparatus and image processing method of thereof | 21 |
| 9 | US20150317772A1 | Systems and methods for chroma noise reduction | 21 |
| 10 | US20220060619A1 | Image processing based on object categorization | 20 |
Citation counts are drawn from within this searched corpus and favour older records; treat them as a signal of influence on later filers, not of current commercial importance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the numbers mean for a filing decision
Three patterns stand out once volume, citations, and IPC composition are read together.
The core claim space is pictorial communication, not AI
H04N and G06T together account for the large majority of records. G06N — AI-model computing — appears in only 14 records, meaning most of the field's claim density is still in classical pipeline stages: demosaicing, tone mapping, defect correction.
Old tone-mapping art still anchors the field
The most-cited record, on local tone mapping, carries 226 citations — far ahead of the next tier. New filings in HDR and tone reproduction are almost certainly citing back to this lineage, which narrows the room for a genuinely novel independent claim in that sub-area.
Filing strategy is US-first with a thin international tail
The United States receiving office accounts for the bulk of records, with Europe and the PCT route a distant second and third. South Korea and China each register only a handful — a gap worth checking against any assignee's actual manufacturing or sales footprint in Asia.
The largest assignees have gone quiet in the latest year
Recent-year momentum among the historically largest filers is flat at best and negative for at least one major assignee. That slowdown at the top, combined with a 2024 peak already behind us, points to a maturing claim landscape rather than an active filing race.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to image signal processor technology landscape, with the prior art for and against each one.
Who is filing, and who has stopped
Ranking tables elsewhere on this page show the full assignee order. The pattern worth flagging here is what recent-year momentum reveals that a cumulative count does not.
A handful of large device and chip makers anchor the corpus
The assignee ranking is led by the kind of companies that ship camera hardware at volume — mobile chipset vendors, device makers, and memory/sensor suppliers. Their historical filing volume is what built the H04N and G06T concentration seen in the IPC breakdown.
Even the most active filer isn't accelerating
The assignee with the most recent-year activity still shows 0% year-over-year change, and several other historically prominent filers show none at all in the latest year. That is a flat signal across the board, not just at one company.
Co-filing is limited and mostly individual-to-individual
Only 10 co-assignee pairs appear in the corpus, and the strongest pairing links two named individuals rather than two corporations. Cross-company joint filing is not a major feature of this field yet.
| Assignee | Recent year | YoY |
|---|---|---|
| Qualcomm Incorporated | 1 | 0% |
| GoPro, Inc. | 0 | — |
| Apple Inc. | 0 | — |
| Samsung Electronics Co., Ltd. | 0 | — |
| SK Hynix Inc. | 0 | — |
| Intel Corporation | 0 | -100% |
| Carl Zeiss AG | 0 | — |
| Huawei Technologies Co., Ltd. | 0 | — |
Where to take this from here
The landscape points to a field with occupied core claims and a quiet top tier — the next steps depend on whether you're filing, licensing, or designing around.
Check freedom-to-operate against the top-cited lineage
Before filing in tone mapping, highlight recovery, or defect correction, run a citation check against the most-cited records in this set — they anchor large citation trees that newer filings tend to sit inside.
Run a freedom-to-operate check in EurekaWatch for re-acceleration among quiet top filers
Flat or negative year-over-year momentum among historically large assignees can reverse quickly once a product cycle turns. Set an alert on the largest filers rather than assuming the slowdown is permanent.
Set an assignee monitor in EurekaScope claims toward the thinner IPC branches
G06N-tagged AI-model claims and B60R vehicle-specific overlaps carry far fewer records than the core H04N/G06T space — a more defensible entry point for new filings.
Explore white space in EurekaCommon questions on image signal processor patents
This landscape identifies 235 patent families published between 2015 and mid-2026 that match image signal processor pipeline claims under IPC classes H04N23/80, H04N25/00, and G06T5. That figure covers pipeline-stage work such as HDR fusion, demosaicing, noise reduction, and on-sensor processing rather than the broader imaging or display fields. Because publication lags filing by roughly 18 months, the true 2025 and 2026 totals are understated in any trend line pulled today.
The assignee ranking is led by major mobile chipset and device makers, with memory and sensor suppliers also holding meaningful positions. Recent-year momentum data shows several of these same top historical filers with flat or negative year-over-year activity, which suggests leadership by cumulative volume does not currently translate into the fastest-growing filer. Anyone assessing competitive risk should weight both the ranking and the momentum figures rather than the ranking alone.
Filings grew from 9 in 2017 to a peak of 26 in 2024, passing through 18 at the 2022 midpoint. That is a decade-long steady build rather than a sudden surge, and the growth curve has flattened since the 2024 peak. The partial 2026 count should be read as an undercount given normal publication delay.
US9787922B2, assigned to Apple and published in 2017, covers a pixel defect preprocessing component that adjusts patterned defect pixels using an interpolation technique against a gain lookup table before passing corrected data to downstream processing modules. It sits early in this corpus's timeline, ahead of the field's peak filing years, and is narrowly scoped to that specific defect-correction mechanism rather than the ISP pipeline broadly. It does not block general demosaicing, HDR fusion, or noise-reduction work outside that specific gain-table interpolation approach, but anyone building patterned-defect correction should review its claims directly rather than assume clearance.
IPC composition shows the bulk of records concentrated in H04N and G06T, with G06N (AI-model computing) and B60R (vehicle-specific applications) carrying far fewer filings. Joint denoising-demosaicing approaches that combine neural-network methods with classical pipeline stages also appear thinly represented relative to the core corpus. These lower-density branches are a more defensible place to scope a first claim than the heavily-cited tone-mapping and highlight-recovery space.
Research Image Signal Processor Technology Landscape in depth with Eureka
Go past this page: query the whole image signal processor technology landscape corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.