Image Sensor Tag Information Embedding for Frame Order Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sensors face challenges in reconstructing image data in temporal order after high frame rate imaging, as the output order of frames differs from the capture order, making it difficult to use methods like time stamping in headers due to increased header size.
Innovation Solution
An image sensor with a pixel unit, image generating unit, tag information generating unit, and output unit that operates in Slow Movie and Super Slow Movie modes, embedding tag information like SLOW_MODE, FRM_RATE_RATIO, SLOW_NUM, FRM_CNT, FRM_CNT_SS, DUPLI_PIC, and SCN_DET into image data and outputting them through multiple logical channels to ensure correct temporal ordering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If time stamping method is used to record frame order, then frame temporal order can be reconstructed, but header size becomes large
Solution Approach 1:
The patent extracts the frame ordering information from the header and places it in a separate metadata structure. Instead of embedding time stamps or sequence numbers in each frame header, the system uses a compact identification mechanism where frame order information is stored externally, significantly reducing the header size while maintaining the ability to reconstruct temporal order.
Solution Approach 2:
The patent changes the parameter representation from full time stamp values to compact frame identification numbers. By using a simplified indexing system where frames are identified by compact integers rather than full temporal measurements, the header size is reduced while still enabling accurate temporal reconstruction through the relationship between frame IDs and their corresponding time points.
2Productivity
If high frame rate imaging is performed, then more image data is captured, but output order differs from capture order requiring reconstruction
Solution Approach 1:
The patent applies preliminary action by assigning frame identification information at the moment of capture rather than performing complex ordering operations later. Each frame is tagged with its capture sequence number during high-frame-rate acquisition, and this pre-assigned identification enables straightforward temporal reconstruction without requiring complex sorting or reordering algorithms in post-processing.
3Adaptability or versatility
If frames are output through multiple logical channels, then flexible data routing is achieved, but frame order management becomes more complex
Solution Approach 1:
The patent implements a universal frame identification mechanism that works across multiple logical channels. The same frame tagging and identification system is applied regardless of which logical channel the frame traverses, allowing flexible routing through different channels while maintaining consistent order management. The identification system is channel-agnostic and can handle frames from any source or destination.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables the reconstruction of image data in temporal order, allowing for efficient output and processing of frames in both Slow Movie and Super Slow Movie modes, addressing the issue of frame order discrepancy.
Implementation Method 1
a pixel unit configured to generate a pixel signal in response to incident light
Data Source
AI summary
The present disclosure relates to an image sensor and an electronic device capable of outputting image data corresponding to frames of a moving image in such a manner that the image data can be reconstructed in temporal order.An image sensor according to a first aspect of the present disclosure includes: a pixel unit configured to generate a pixel signal in response to incident light; an image generating unit configured to generate image data corresponding to frames of a moving image on the basis of the generated pixel signal; a tag information generating unit configured to generate tag information to be embedded into the generated image data; an output unit configured to output the image data having the tag information embedded therein through at least one of first and second logical channels; and a control unit configured to control the pixel unit, the image generating unit, the tag information generating unit, and the output unit to operate in a Slow Movie mode or in a Super Slow Movie mode. The present disclosure is applicable to an electronic device having an imaging function.