Image Processing Apparatus Singular Value Decomposition Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in reducing processing load while maintaining object recognition accuracy, particularly in applications like mobile bodies where high-resolution images are required for accurate object detection.
Innovation Solution
The implementation of low-rank approximation by singular value decomposition on acquired images, which reduces the processing load by compressing less significant image portions and maintaining essential characteristics for object recognition, thereby suppressing a decrease in accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If low-rank approximation by singular value decomposition is applied to images, then processing load is reduced, but image quality and object recognition accuracy may deteriorate
Solution Approach 1:
The patent changes the parameter of image representation by applying low-rank approximation through singular value decomposition, transforming the original high-dimensional image data into a lower-dimensional representation that retains essential features while reducing processing complexity
Solution Approach 2:
The patent extracts and removes less significant components from the image data by performing low-rank approximation, keeping only the most important features that contribute to object recognition while discarding redundant information that increases processing load
2Productivity
If image resolution is reduced to decrease processing load, then processing speed increases, but object detection accuracy deteriorates
Solution Approach 1:
Instead of simply reducing image resolution, the patent changes the fundamental parameter of image representation by applying low-rank approximation, which reduces the effective dimensionality of the data while preserving the most significant features needed for accurate object detection
3Measurement precision
If high-resolution images are processed to maintain object detection accuracy, then processing load increases, but energy consumption increases
Solution Approach 1:
The patent transforms the energy consumption problem by changing the data representation parameter through low-rank approximation, reducing the amount of data that needs to be processed while maintaining the essential information required for accurate object detection
Solution Approach 2:
The patent extracts and removes redundant information from high-resolution images through low-rank approximation, keeping only the critical features that contribute to object detection accuracy, thereby reducing the energy required for processing
Data Source
AI summary
An image processing apparatus 10 includes an interface 12 configured to acquire an image and a processor 13 configured to perform low-rank approximation by singular value decomposition on the acquired image, and perform object recognition on the acquired image subjected to the low-rank approximation by the singular value decomposition.