Image Processing Device Region-Based Detail Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-resolution and high-frame-rate image processing in content processing systems leads to increased latency due to the high amount of data that needs to be processed, which can hinder immediate reflection of real-world motions in displayed images.
Innovation Solution
An electronic device and content processing system that acquire and process images by dividing them into regions based on measured values, establishing different levels of detail for each region to reduce data, and outputting the compressed image data along with information about the regions, allowing for efficient data reduction and reproduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the resolution or frame rate of captured images is increased to achieve realistic image renderings or highly accurate image processing sequences, then the image quality and processing accuracy are improved, but the amount of data to be processed increases, resulting in increased latency from image capturing to image displaying
Solution Approach 1:
The captured image is divided into multiple regions based on boundary conditions from sensor images (such as depth information). Different levels of detail are established for different regions, allowing important regions to maintain high resolution while less important regions use lower resolution, thereby reducing overall data amount while preserving processing accuracy where needed.
Solution Approach 2:
Different regions of the captured image are assigned different levels of detail based on their importance. Regions containing important features or objects maintain high resolution and detail, while other regions use reduced detail. This local differentiation reduces total data transmission requirements while maintaining processing accuracy in critical areas.
2Illumination intensity
If the resolution or frame rate of captured images is increased to achieve realistic image renderings, then the visual quality is improved, but the amount of data to be transmitted increases, resulting in increased latency
Solution Approach 1:
The image is segmented into regions with different quality requirements. By transmitting only the necessary amount of data for each region based on its importance, overall transmission data volume is reduced while maintaining visual quality in important areas, thereby reducing transmission latency.
Solution Approach 2:
Different regions are assigned different quality levels. Important regions maintain high visual quality with detailed image data, while less important regions use lower quality representations. This selective quality approach reduces total data transmission while preserving visual quality where it matters most.
3Measurement precision
If high-resolution captured images are processed to maintain immediate reflection of real-world motions, then the image detail is preserved, but the processing speed decreases due to the large amount of data
Solution Approach 1:
By segmenting the image into regions of different importance and processing only the necessary detail for each region, the total processing workload is reduced. Important regions receive full processing attention while less important regions use simplified processing, thereby improving overall processing speed while maintaining detail where needed.
Solution Approach 2:
Processing resources are allocated locally based on region importance. High-detail processing is applied only to important regions while lower-detail processing is used elsewhere. This selective processing approach maintains image detail in critical areas while improving overall processing throughput and speed.
Data Source
AI summary
An image processing device acquires a captured image and a depth image. A region dividing section divides the plane of the captured image into regions by determining distance values indicated by the depth image with respect to a preset threshold value (region-divided image). A compressing section generates and outputs data of a captured image having different levels of detail in the respective regions, with respect to the resolution and the number of gradations representing pixel values of the captured image, according to the result of the region division.