Fast Watermark Detection in Portable Computing Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Portable computing devices like mobile phones face significant strain in multimedia signal processing, requiring enhanced architectures and computing methods to efficiently implement various functions across different hardware and software platforms, with existing distributed computing schemes not directly translating to mobile phone architectures.
Innovation Solution
A reader system with a library that adapts to device capabilities and business model parameters, utilizing reader modules for content identification, incorporating signal filtering, watermark detection, and fingerprinting to optimize resource use on both the device and network, with modules like fast watermark detection and fingerprinting to efficiently identify content items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed computing schemes are implemented to break complicated software tasks into modules and distribute execution across networked processors, then computing efficiency is improved, but device complexity increases due to the need for new computing schemes and architectures for mobile phone environments
Solution Approach 1:
The patent segments the content identification task into multiple reader modules (watermark detection module, fingerprint module, filter module) that can be selectively executed. The library divides the overall processing into discrete functional units that can be distributed across different execution environments (handset processor, server, peer device), allowing complex tasks to be broken down without increasing overall system complexity.
Solution Approach 2:
The patent creates a universal reader library that can execute the same reader modules across different platforms and devices (handsets, servers, peer devices). The modules are designed to be platform-independent, allowing the same content identification logic to run universally across diverse computing environments without requiring device-specific implementations.
2Measurement precision
If multiple reader modules are selected and executed based on device capabilities and business model parameters, then content identification accuracy is improved, but resource consumption increases due to the need to process multiple module configurations
Solution Approach 1:
The patent implements dynamic module selection where the library evaluates device capabilities, content characteristics, and business model parameters at runtime to determine which reader modules to execute. This dynamic adaptation allows the system to use more modules when resources are abundant and fewer modules when resources are constrained, optimizing the balance between identification accuracy and resource consumption.
Solution Approach 2:
The patent performs preliminary evaluation of device capabilities and content characteristics before selecting reader modules. The library assesses available resources, content type, and identification requirements in advance, then pre-selects the appropriate module configuration. This preliminary action prevents unnecessary module execution, reducing resource consumption while maintaining identification accuracy.
3Speed
If fast watermark detection is implemented to quickly detect the presence of watermarks and focus resources on likely content, then processing speed is improved, but detection precision may be reduced due to the simplified detection approach
Solution Approach 1:
The patent segments the detection process into two stages: a fast preliminary watermark presence detection stage, and a more precise content identification stage. The fast detection module quickly screens for watermark presence using simplified algorithms, then only content that passes this screen proceeds to the more precise fingerprinting and full watermark detection modules, maintaining both speed and precision.
Solution Approach 2:
The patent applies different detection qualities to different content regions and types. The fast detection uses simplified algorithms for initial screening, while the precise detection applies sophisticated algorithms only to regions and content types where high precision is critical. This local quality differentiation maintains overall detection precision while improving processing speed.
Data Source
AI summary
This disclosure describes a distributed reader architecture for a mobile computing device such as cellular telephone handset. This architecture includes a reader library that reads device capabilities and business model parameters in the device, and in response, for selects an appropriate configuration of reader modules for identifying a content item. The reader modules each perform a function used in identifying a content item. The modules are selected so that the resources available on the device and in remote devices are used optimally, depending on available computing resources on the device and network bandwidth. One example of a reader module is a fast watermark detection module that quickly detects the presence of a watermark, enabling resources to be focused on portions of content that are most likely going to lead to successful content identification. A watermark signal structure for fast watermark detection is comprised of a dense array of impulse functions in a form of a circle in a Fourier magnitude domain, and the impulse functions having pseudorandom phase. Alternative structures are possible.