Image Segment Statistics Using Mantissa-Exponent Memory Splitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in handling large amounts of data due to limitations in integer and floating-point representations, leading to increased computational footprint and resource consumption without maintaining accuracy.
Innovation Solution
The image is split into segments, with statistical values represented as a combination of mantissa and exponent stored in separate memory portions, reducing memory consumption by using a formulaic representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional integer or floating-point formats are used to represent statistical values for image segments, then measurement accuracy is maintained, but memory consumption and computational footprint increase significantly
Solution Approach 1:
The statistical value is segmented into two distinct parts: a mantissa portion and an exponent portion. The mantissa stores the significant digits while the exponent stores the magnitude information. This segmentation allows the system to represent large statistical values with reduced memory footprint compared to traditional fixed-width integer or floating-point formats, while preserving measurement accuracy through the combined representation.
Solution Approach 2:
The patent changes the parameter representation from traditional integer or floating-point formats to a custom formulaic representation using mantissa and exponent components. This parameter transformation enables the system to adapt the precision and range of statistical values dynamically, reducing memory consumption while maintaining the required measurement accuracy for image processing applications.
2Loss of information
If the amount of data associated with an image increases, then more comprehensive statistical analysis is possible, but the computational footprint expands beyond reasonable limits
Solution Approach 1:
By segmenting the statistical value representation into mantissa and exponent portions, the system can process comprehensive statistical data from image segments without requiring proportional increases in memory allocation. Each segment's statistics are encoded efficiently, allowing complete statistical analysis across the entire image while maintaining a constrained computational footprint.
Solution Approach 2:
The patent introduces a new dimensional approach to data representation by using a formulaic structure with separate mantissa and exponent fields. This dimensional change in data organization allows the system to accommodate increased statistical data volume without linearly increasing the computational footprint, as the formulaic representation compresses the information density.
3Quantity of substance
If the computational footprint is maintained at reasonable levels, then resource consumption is controlled, but the effective amount of data that can be processed is reduced
Solution Approach 1:
By changing the parameter representation to a formulaic structure with mantissa and exponent components, the system achieves better data processing capacity within the same computational footprint. The parameter transformation allows more statistical information to be encoded in the same memory space, effectively increasing data processing capacity without expanding the computational resource requirements.
Data Source
AI summary
Aspects of the subject disclosure may include, for example, splitting an image that is obtained from a video adaptor into a plurality of segments, for each segment of the plurality of segments, generating a value that is representative of data of the segment, the value resulting in no loss in an accuracy of the data, storing a first part of the value in a first portion of a memory, and storing a second part of the value in a second portion of the memory, the second part being different from the first part. Other embodiments are disclosed.