Complex-Valued Image Data Compression via Magnitude-Phase Quantization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for compressing and transmitting image data with phase components require the entire data set to be received before processing can begin, leading to significant wait times due to large data volumes and low network bandwidth.
Innovation Solution
The method involves converting complex-valued image data into scaled coordinate pairs in a magnitude-phase plane, quantizing these values, and transmitting them in decreasing bit significance order, allowing for progressive processing and transmission, utilizing hardware parallel processing devices for efficient data handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If complex-valued image data is transmitted in traditional format, then data completeness is ensured, but transmission time and processing delay increase significantly
Solution Approach 1:
The complex-valued image data is segmented into magnitude and phase components, which are then independently quantized and transmitted. This segmentation allows the data to be processed and transmitted more efficiently, reducing transmission time while maintaining essential information through separate encoding of magnitude and phase data.
Solution Approach 2:
The magnitude and phase components are pre-quantized and encoded before transmission. By performing quantization and bit ordering in advance at the transmitter side, the data is prepared for efficient transmission and progressive reconstruction, reducing the time required during actual transmission and processing.
2Productivity
If data compression is applied to complex image data, then transmission efficiency improves, but processing complexity increases
Solution Approach 1:
The data is divided into magnitude and phase components that can be independently processed and quantized. This segmentation simplifies the compression process by allowing separate handling of each component, reducing overall processing complexity while improving transmission efficiency through targeted quantization of each segment.
Solution Approach 2:
The patent applies quantization to transform continuous magnitude and phase values into discrete representations. This parameter change from continuous to discrete domain enables efficient compression and reduces data volume for transmission, while the standardized quantization process keeps processing complexity manageable.
3Measurement precision
If full precision complex data is transmitted, then image quality is maintained, but network bandwidth requirements increase
Solution Approach 1:
By separating complex data into magnitude and phase components and quantizing them independently, the patent reduces the total data volume required for transmission. The segmentation allows for optimized bit allocation to each component, maintaining necessary precision while reducing overall data quantity transmitted over the network.
Solution Approach 2:
The patent extracts only the essential magnitude and phase information from the complex-valued data, discarding redundant representations. This extraction process reduces data volume by transmitting only the critical components needed for image reconstruction, rather than transmitting the full complex data structure.
Data Source
AI summary
The present disclosure generally relates to techniques for processing complex-valued array data representing an image. The techniques may include obtaining an electronic representation of an array of complex numbers representing an image, converting the array of complex numbers to an array of scaled coordinate pair values in a magnitude-phase plane, replacing each coordinate pair value with data representing a respective nearest node in a quantized magnitude-phase plane, such that an array of scaled quantized coordinate value pairs is produced, arranging into a sequence of bit values ordered according to decreasing bit significance, from most-significant bit values to least-significant bit values, the scaled quantized coordinate value pairs, and transmitting the sequence of bit values to a receiver, such that the receiver rearranges and rescales the sequence of bit values and obtains the image represented by the array of complex numbers.