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 →This landscape tracks patent families at the intersection of CMOS image sensor fabrication and process automation — wafer fab automation, packaging automation, hybrid bonding automation and process control — filed against IPC classes covering semiconductor devices (H01L27/146), general semiconductor processing (H01L21/00) and programmable controllers (G05B19/418). The 86 families identified span publication years 2015 through the partial 2026 cut-off, with filing activity concentrated in the mid-2010s.
Because publication lags filing by roughly 18 months, the 2025 and 2026 figures in any trend line understate true filing activity; treat the most recent one to two years as a floor, not a ceiling.
Pick a task. Every answer cites the patents behind it.
Two views of the same 86-family corpus: how filing volume has moved year over year, and which IPC subclasses carry the claim weight.
Filings hit 11 in 2017, the high point of the entire window, then fell toward the 2022 midpoint of 3. That pattern — a sharp early peak followed by a decade-long plateau below it — points to a technology area where the foundational claim space filled early and later entrants are filing incrementally rather than opening new ground.
H01L (semiconductor devices) covers 78 of 86 records, and H04N (pictorial communication) covers 34, but G05B (control and regulating systems) covers only 7. The imaging device itself is heavily claimed; the automation and process-control layer that manufactures it is comparatively open. G06T (image data processing) matches G05B at 7, and A22C — an unrelated meat and fish processing class picked up by the search string's broad TACD terms — appears at 5, a reminder to sanity-check outlier classes before citing them as signal.
Shares are the percentage of the 86 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
This page is one run against one query. Ask Eureka your own question about cmos image sensor process automation and every answer comes back with the patent numbers behind it.
Try EurekaDisclosed is a fault tolerant CMOS image sensor that includes circuitry for identifying defective pixels and masking them during image generation. Masking may involve, in one example, replacing the output of a given pixel with an average of the output of surrounding non-faulty pixels. Thus, while image sensors may be fabricated with some number of faulty pixels, the images produced by such sensors will not have undesirable bright or dark spots.Filed by STMicroelectronics S.r.l., granted 2003-09-09. With 100 citations inside this corpus, it is the single most-referenced record and a mandatory freedom-to-operate check for any defect-masking or pixel-correction design.


| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US6618084B1 | Pixel correction system and method for CMOS imagers | 100 |
| 2 | US5889277A | Planar color filter array for CCDs with embedded color filter elements | 89 |
| 3 | US4901129A | Bulk charge modulated transistor threshold image sensor elements and method of making | 68 |
| 4 | US20020048837A1 | Fabrication of a high-precision blooming control structure for an image sensor | 47 |
| 5 | US20170040357A1 | Image sensor and method for fabricating the same | 41 |
| 6 | US9147708B2 | Solid-state image sensor and camera | 35 |
| 7 | US6188056B1 | Solid state optical imaging pixel with resistive load | 33 |
| 8 | US20040095488A1 | Pixel correction system and method for CMOS imagers | 26 |
| 9 | EP0917358A1 | Pixel correction system and method for CMOS imagers | 25 |
| 10 | US9276027B2 | Solid-state image sensor and camera | 22 |
Citation counts are drawn from the searched corpus only and favour older filings; treat them as a measure of influence on later claims, not of current commercial relevance.
Each row carries its publication number; clicking a row searches Eureka by that number.
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 →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 →Three signals from this corpus matter more than the raw family count when deciding where to file next.
Filing volume ran from a peak of 11 in 2017 down through a midpoint of 3 in 2022. Recent-year momentum tables show zero new filings in the latest year across every tracked assignee, including established names. Discount the very last year or two for publication lag, but the multi-year decline predates that effect.
Semiconductor device structure (H01L) accounts for the overwhelming majority of records, while control and regulating systems (G05B) accounts for only 7. Automation logic that governs wafer fab or hybrid bonding steps is far less densely claimed than the sensor structures it produces.
US6618084B1 carries 100 citations against a corpus of 86 families — more citations than there are records in the set. Any defect-correction or pixel-masking filing in this space should expect examiners to raise it.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to cmos image sensor process automation, with the prior art for and against each one.
| Assignee | Co-assignee | Shared families |
|---|---|---|
| Semiconductor Manufacturing International (Beijing) Corporation (SMIC Beijing) | Semiconductor Manufacturing International (Shanghai) Corporation (SMIC Shanghai) | 5 |
| Soitec | MAZURE CARLOS | 1 |
| Soitec | BOURDELLE KONSTANTIN | 1 |
Only 3 co-assignee pairs appear across the corpus, the strongest linking two SMIC entities on 5 shared records. Cross-company collaboration on process automation claims is the exception, not the norm, in this data set.
Recent-year momentum shows every tracked assignee, from foundries to equipment makers, at zero new filings in the latest year. That flatness across the board — not a single company pulling ahead — is itself the finding.
Canon, TSMC, STMicroelectronics, JBT Marel, Soitec and SMIC's Beijing entity all show zero filings in the latest tracked year, with JBT Marel down 100% year over year. This is consistent with the broader trend decline rather than a company-specific pullback.
The strongest co-assignee link — SMIC's Beijing and Shanghai manufacturing entities on 5 shared records — reflects internal corporate structure more than external collaboration. Soitec appears twice as a co-assignee, each time on a single record with an individual inventor.
The United States accounts for 48 of the tracked records, more than the next four receiving offices — EPO, WIPO, China and Canada — combined. A US-first filing strategy is the norm for this technology area, with EPO as the clear secondary venue at 13.
| Assignee | Recent year | YoY |
|---|---|---|
| Canon Inc. | 0 | — |
| Taiwan Semiconductor Manufacturing Company (TSMC) | 0 | — |
| STMICROELECTRONICS INC | 0 | — |
| JBT Marel | 0 | -100% |
| Soitec | 0 | — |
| Semiconductor Manufacturing International (Beijing) Corporation (SMIC Beijing) | 0 | — |
| Semiconductor Manufacturing International (Shanghai) Corporation (SMIC Shanghai) | 0 | — |
| OmniVision Technologies | 0 | — |
The trend and assignee data point to specific next steps depending on what you are trying to decide.
With US6618084B1 carrying 100 citations in this corpus, any pixel-masking or defect-correction filing should map its claims against it before drafting.
Check claim overlap in EurekaG05B and G06T each cover only 7 of 86 records against 78 for H01L. A first claim drafted around automated process-control logic for hybrid bonding or wafer-level calibration is filing into comparatively open space.
Explore the gap in EurekaZero filings in the latest year across every tracked assignee is partly a lag artefact. Re-run this trend in 12–18 months before concluding the space is fully dormant.
Set a monitoring alert in EurekaFiling activity peaked in 2017 at 11 records and declined toward a midpoint of 3 in 2022, with recent-year momentum showing zero new filings across every tracked assignee. Some of the most recent apparent drop-off is a publication-lag artefact, since patents typically publish around 18 months after filing, so the last one to two years will always look thinner than they eventually turn out to be. Taken together, the multi-year trend before the lag window suggests this is a mature claim space rather than a growing one, and new entrants should expect to compete against art from the 2015–2019 period rather than a rapidly shifting frontier.
US6618084B1, titled 'Pixel correction system and method for CMOS imagers' and granted to STMicroelectronics S.r.l. in 2003, is the most-cited record in this corpus with 100 citations. It covers a fault-tolerant sensor architecture that identifies defective pixels and masks them during image generation, commonly by averaging surrounding non-faulty pixel outputs. Any new filing involving defect masking, pixel correction or blooming control should review this patent's claims closely, since its citation count exceeds the total number of families in this entire landscape.
78 of the 86 tracked families are classified under H01L, covering semiconductor device structure, while only 7 fall under G05B, which covers control and regulating systems. This reflects where the search terms and the underlying industry attention concentrate: patent applicants have historically claimed the physical sensor and fabrication structure far more heavily than the automation logic that runs the fabrication line. It does not mean automation technology is unpatented elsewhere — process-control patents may simply be classified under different IPC codes outside this search's scope — but within this dataset, control-layer claims are comparatively sparse.
The recent-year momentum data shows every tracked assignee — including Canon, TSMC, STMicroelectronics, Soitec, JBT Marel and SMIC's Beijing entity — at zero filings in the latest tracked year. No single company is currently accelerating ahead of the others; the flat momentum across the board mirrors the broader decline in the filing trend since 2017. Co-filing is also rare, with only 3 co-assignee pairs identified across all 86 families, the strongest being between two SMIC manufacturing entities rather than an external partnership.
The clearest gap sits in process-control and image-processing classifications: G05B and G06T each cover only 7 of 86 records compared with 78 for device-structure class H01L. Specific under-claimed branches include hybrid bonding process-control loops, automated wafer-level defect classification logic, and programmable-controller integration for packaging automation. These are areas where the manufacturing process itself, rather than the resulting sensor structure, is the subject of the claim — a distinction worth building a first claim around if the goal is to avoid the densest prior art.
Go past this page: query the whole cmos image sensor process automation 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.