Image Conversion Apparatus Iterative Coordinate Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image conversion methods require significant computational resources for inverse projection transformation due to the need for multiple multiplication processes, leading to high load and reduced image quality, especially in real-time processing of moving images, and they often require large buffer memories for block-based image coding.
Innovation Solution
An image conversion apparatus and method that uses iterative addition and subtraction processes with a predetermined constant to simulate multiplication results, reducing the need for multiple multipliers and enabling efficient coordinate conversion without interpolation between neighbor pixels, thus minimizing computational load and improving image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple multiplication processes are used for inverse projection transformation, then coordinate conversion accuracy is improved, but computational load increases and processing speed decreases
Solution Approach 1:
The patent changes the mathematical operation parameters from multiplication to addition/subtraction. By transforming the inverse projection transformation equations to use only addition and subtraction operations, the computational complexity is reduced while maintaining the same coordinate conversion accuracy, thus resolving the contradiction between precision and processing speed.
Solution Approach 2:
The patent substitutes the mechanical multiplication operation with simpler addition and subtraction operations. This replacement of computational mechanisms allows for faster processing without sacrificing the mathematical accuracy of coordinate conversion, directly addressing the speed-accuracy tradeoff.
2Speed
If hardware with multiple multipliers is used for inverse projection transformation, then calculation speed is improved, but device complexity increases
Solution Approach 1:
The patent extracts and eliminates the multiplication operation from the inverse projection transformation calculation. By removing the need for multipliers and replacing them with addition/subtraction operations, the hardware complexity is significantly reduced while maintaining calculation speed through simpler arithmetic units.
Solution Approach 2:
The patent replaces expensive and complex multiplier hardware with simpler, more economical addition and subtraction circuits. This substitution uses cheaper computational resources to achieve the same functional result, reducing device complexity and cost.
3Measurement precision
If large buffer memories are used for block-based image coding, then image coding accuracy is improved, but memory requirement increases
Solution Approach 1:
The patent segments the image processing into smaller blocks that can be processed independently. By dividing the image into manageable blocks and processing them sequentially with reduced buffer requirements, the system achieves adequate coding accuracy without needing large buffer memories, thus resolving the contradiction between accuracy and memory capacity.
Data Source
AI summary
An image conversion apparatus calculates, based on a first value for obtaining first coordinate values in a second image before first conversion, which correspond to coordinate values of one pixel in a first image after first conversion, a second value for obtaining second coordinate values in the second image, which correspond to coordinate values of a pixel adjacent to the one pixel in the first image. The apparatus converts the second coordinate values into third coordinate values for second conversion of converting a third image into the second image and converts the third image into the first image. In the calculation of the second value, addition or subtraction using a constant and a result of the calculation is iteratively executed for sequentially outputting values corresponding to results of multiplication of coordinate values of each pixel in the first image and the constant.