HDR Image Data Compression via Color Channel Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High dynamic range (HDR) image data transmission requires high-bandwidth communication interfaces, leading to increased hardware costs and power consumption, and existing compression methods suffer from reduced efficiency due to significant level differences between adjacent pixels.
Innovation Solution
An electronic device with an image sensor having different exposure settings for two sensor pixel sets, which splits and compresses raw image data into channel data, generating adjacent data patterns for efficient compression and transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high-bandwidth communication interface is used to transmit HDR image data, then transmission quality is improved, but hardware cost and power consumption increase
Solution Approach 1:
The raw image data is segmented into multiple channels based on color information (e.g., R, G, B channels). Each channel is compressed independently using run-length encoding, which groups adjacent pixels of the same color. This segmentation allows for more efficient compression compared to processing all pixels uniformly, thereby reducing the required transmission bandwidth and power consumption while maintaining image quality
Solution Approach 2:
The patent changes the parameter of data representation by organizing pixel data according to color channels and applying run-length encoding. This transforms the raw image data into a compressed format that requires fewer bits for transmission, directly reducing the bandwidth requirement and associated power consumption of the communication interface
2Reliability
If high-bandwidth communication interface is used to transmit HDR image data, then transmission quality is improved, but hardware cost increases
Solution Approach 1:
The image data is divided into color channels and further segmented into sequences of adjacent pixels within each channel. This segmentation enables run-length encoding to be applied effectively, compressing the data and allowing transmission over lower-bandwidth, less expensive communication interfaces while maintaining transmission quality
Solution Approach 2:
The patent creates a compressed representation (copy) of the original image data through run-length encoding. This compressed copy contains the same essential information but in a more compact form, allowing transmission over cheaper interfaces without losing transmission quality
3Productivity
If conventional compression is applied to HDR image data with significant level differences between adjacent pixels, then compression speed is improved, but compression efficiency deteriorates
Solution Approach 1:
The patent changes the organization parameter of the data by grouping adjacent pixels of the same color within each color channel. This reorganization creates long runs of identical values that are highly compressible using run-length encoding, significantly improving compression efficiency compared to conventional methods that process pixels in their original spatial arrangement
Solution Approach 2:
By segmenting the image data into color channels and then into sequences of adjacent pixels within each channel, the patent creates structures that are highly amenable to run-length encoding. This segmentation approach maintains compression speed while dramatically improving compression efficiency for HDR image data with significant level differences between adjacent pixels
Data Source
AI summary
Various embodiments of the present invention relate to an electronic device comprising: a processor; an image sensor that includes a first sensor pixel set and a second sensor pixel set; and a control circuit that is electrically connected to the image sensor, wherein the control circuit can be configured so as to obtain raw image data by using the image sensor in which the exposure of the first sensor pixel set and the exposure of the second sensor pixel set are configured differently, divide the raw image data into first channel data corresponding to a first color and second channel data corresponding to a second color, generate first channel compression data that are compressed on the basis of the first channel data such that at least a part of data obtained using the first sensor pixel set adjoins and at least a part of data obtained using the second sensor pixel set adjoins, generate second channel compression data that are compressed on the basis of the second channel data such that at least a part of the data obtained using the first sensor pixel set adjoins and at least a part of the data obtained using the second sensor pixel set adjoins, and transmit the compressed first channel compression data and the compressed second channel compression data to the processor. In addition, among the various embodiments of the present invention other embodiments may be possible.