Structured Data Stream for Computational Photography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The computational photography system faces inefficiencies in image processing due to low processing efficiency and poor compatibility and consistency, primarily because it relies on transformation-based solutions that require multiple queries for characteristic information, leading to cumbersome and inconsistent data formats across different modules.
Innovation Solution
An image processing method that sets a specific data stream format to include both image data and characteristic data, allowing for direct extraction and processing without the need for repeated queries, thereby improving efficiency and consistency by organizing characteristic information in a structured format within the data stream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If transformation-based solutions are used in the computational photography system, then the system can process images, but the image processing efficiency is relatively low due to requiring multiple queries for characteristic information
Solution Approach 1:
The patent applies preliminary action by embedding all characteristic information (exposure parameters, lens parameters, sensor parameters, etc.) into the image data before processing. This allows the computational photography system to access all necessary parameters in one operation rather than requiring multiple sequential queries, thereby improving processing efficiency and reducing time loss.
2Adaptability or versatility
If transformation-based solutions are used in the computational photography system, then the system can process images, but the compatibility and consistency are relatively poor across different modules
Solution Approach 1:
The patent applies merging by combining image data and characteristic information into a unified data structure. This integration ensures that all modules access data in a consistent format, improving compatibility and consistency across the computational photography system while reducing the complexity associated with handling multiple disparate data formats.
3Loss of information
If multiple queries are required for characteristic information, then comprehensive data can be obtained, but the processing becomes cumbersome and inefficient
Solution Approach 1:
The patent applies preliminary action by pre-assembling all characteristic information (exposure, lens, sensor parameters, etc.) with the image data before processing. This eliminates the need for multiple separate queries, maintaining complete information availability while significantly simplifying the operational process and improving ease of use.
4Ease of manufacture
If different data formats are used across modules, then each module can optimize its data structure, but the overall system consistency deteriorates
Solution Approach 1:
The patent applies universality by designing a unified data structure that can accommodate various types of characteristic information (exposure parameters, lens parameters, sensor parameters, etc.) in a consistent format. This universal structure allows different modules to access and process data uniformly, maintaining system consistency while still enabling each module to optimize its processing logic.
Data Source
AI summary
The present application proposes an image processing method, a device and a storage medium. The image processing method includes the following steps: obtaining first image data; determining or generating a target data stream according to a data stream format; and performing image processing on the first image data according to the target data stream. In the present application, by setting the data stream format, the target data stream based on the data stream format can include the first image data and/or characteristic data of the first image data, and the image processing on the first image data can be implemented according to the target data stream, improving the image processing efficiency as well as the compatibility and consistency of a computational photography system.