Obviousness

The following analysis of obviousness for US patent 12250452 is based on the prior art references explicitly described within the patent itself.

Claimed Invention (Key Features from Abstract and Summary of US12250452):
The patent US12250452 describes a computerized method for real-time assessment of picture quality. Key features include:

1. On-board combining of a plurality of quality indicators (QIs) from current and previous image frames.
2. Computation and use of a "confidence level" for at least one of these quality indicators.
3. Using a processor to determine if photo quality is acceptable based on this combining.
4. Taking differential action (e.g., capturing the image if acceptable, providing suggestions, controlling camera parameters).
5. Crucially, the weight of one quality indicator dynamically takes into account data from other quality indicators, including their values, weights, confidence levels, and their previous values, weights, and confidence levels. For example, if device shake or camera focus QIs are poor, the aesthetic QI might be disregarded.
6. Real-time control of lens module parameters (e.g., shutter speed, ISO, aperture, focus point) based on quality indications to achieve better overall quality.
7. Automatic image capture when QI thresholds are met, even without a user pressing the shutter button.
8. Device movement gestures for control (e.g., shutter press or delete).
9. Evaluation of video quality using continuous frames and quality indicators.

Prior Art References (as described in US12250452):

• US 20130155474: Teaches providing feedback prior to capturing an image of a paper document, with instructions for adjusting measured parameters. It allows combining parameter values into a total quality score that must exceed a threshold for image capture. It also suggests holding the camera steadier to avoid blur.
• WO 2006040761/US20070195174: Describes a system that defines scene dynamics (e.g., image motion speed, subject motion speed) and allows for user-settable relative weights for attributes when computing a total image grade.
• US20090278958: Teaches basing the scoring of a current image on scores from previously captured images to avoid redundant calculations.

Obviousness Analysis under 35 U.S.C. § 103:

A person having ordinary skill in the art (POSITA) in image processing or digital photography would be motivated to improve the accuracy, robustness, and automation of real-time picture quality assessment systems.

Combination 1: US 20130155474 + WO 2006040761/US20070195174 + US20090278958

• Motivation to Combine:

  • A POSITA aiming to enhance the real-time feedback and capture threshold system of US 20130155474 would recognize the benefit of more nuanced control over the "total overall quality score."
  • This would naturally lead to incorporating the concept of weighting different quality attributes, as taught by WO 2006040761/US20070195174, to allow certain aspects (e.g., focus) to contribute more or less to the final score than others. While WO 2006040761/US20070195174 specifies "user-settable" weights, a POSITA would continually seek to automate or optimize such settings.
  • Furthermore, a POSITA would be motivated to leverage historical data to make more informed decisions, not just for computational efficiency (as in US20090278958), but to gauge trends or detect anomalies in quality indicators over a sequence of frames.

• Combined Teaching vs. US12250452:

  • This combination would teach a system that:
    • Combines multiple quality parameters into a total score.
    • Uses previous image data for current scoring.
    • Applies (user-settable) weights to different quality attributes.
    • Determines if the total quality is acceptable against a threshold for image capture.
    • Provides feedback or suggestions to the user for improvement (e.g., addressing blur).

• Elements of US12250452 Not Rendered Obvious by this Combination:
  Despite the combined teachings, several key distinguishing features of US12250452 are not explicitly taught or rendered obvious by this combination of prior art snippets:

  1. Confidence Level for Quality Indicators: None of the cited prior art references describe computing or utilizing a "confidence level" for individual quality indicators. US12250452 explicitly defines and uses such a confidence level (C i (t)), often derived from statistical analysis (e.g., normal distribution of probability P i (t) over time) or by detecting inconsistencies between different sensor inputs (e.g., lens module focus report vs. independent blur detection). This concept addresses the reliability of the QIs themselves.
  2. Dynamic Interdependency and Adjustment of QI Weights Based on Confidence Levels: US12250452 teaches that "the weight of one indicator will take into account data from other quality indicator/s e.g. their quality indicator value, weight, confidence level...". This is a crucial distinction from the "user-settable relative weight" of WO 2006040761/US20070195174. The patent specifically illustrates scenarios where a low confidence in one QI (e.g., device shake or focus) can cause the system to disregard or dynamically reduce the weight of another QI (e.g., aesthetic quality), even if the user had initially prioritized it. This dynamic, confidence-driven adjustment goes beyond merely applying static or manually configured weights.
  3. Real-time Automatic Control of Lens/Sensor Module Parameters: US12250452 describes using quality indications to directly control lens module parameters such as shortening shutter speed by increasing ISO or changing aperture, or adjusting focus points, to achieve better total QI. While US 20130155474 mentions "adjusting at least one measured parameter" and providing instructions for user action, it does not disclose automatic, in-camera adjustment of photographic parameters based on real-time quality analysis and sensor fusion.
  4. Automatic Image Capture Without User Shutter Press: Embodiment 8 of US12250452 details a system for automatic image capturing once a configured total quality indicator meets a minimum threshold, even without the user pressing a shutter button. The prior art describes waiting for a threshold to be met after a user has initiated a capture attempt (e.g., for a document).
  5. Device Movement Gestures for Control: Embodiments 12 and 13 introduce using device movement gestures (e.g., shake) for shutter press indication or to delete a picture. This user interface innovation is not suggested by the provided prior art.
  6. Evaluation of Video Quality: Embodiment 11 and its supporting description in US12250452 discuss using the quality ranking system over continuous frames to evaluate video quality, potentially with different weight functions for different frames. This application to video quality assessment is not covered by the described prior art.

Conclusion:
Based solely on the provided descriptions of prior art in US12250452, the key inventive aspects related to the computation and integration of "confidence levels" for quality indicators, the dynamic and inter-dependent adjustment of QI weights based on these confidence levels and other QI values, and the real-time automatic control of camera parameters to improve quality, are not rendered obvious. The prior art teaches elements of image quality assessment, feedback, and parameterized weighting, but it lacks the sophisticated, self-aware (via confidence levels) and dynamically adaptive system for optimizing image capture as claimed in US12250452. Therefore, a POSITA would not have been motivated to combine these references to arrive at these specific distinguishing features of the patent with a reasonable expectation of success, without the benefit of hindsight.