Integral Image Buffer Truncation for Fast Region Sum Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for image processing and pattern recognition require large buffer capacities to hold accumulated image information, leading to increased circuit scale and processing load, especially when calculating the sum total value of pixels in rectangular regions.
Innovation Solution
The method involves inputting multi-dimensional array information, calculating accumulated values, and holding them in a buffer with a predetermined bit size, where overflow values are truncated to fit within the buffer capacity, allowing for efficient storage and calculation without increasing circuit scale or processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If accumulated image information is stored in a buffer with sufficient bit capacity to prevent overflow, then measurement precision is improved, but device complexity and buffer capacity requirements increase
Solution Approach 1:
The patent changes the parameter of bit precision from a fixed large value to a dynamically determined smaller value. By calculating the actual maximum accumulated value and determining the minimum required bit precision, the system reduces buffer capacity from worst-case assumptions to actual needs, resolving the contradiction between precision and device complexity
Solution Approach 2:
Instead of allocating full buffer capacity for worst-case scenarios, the patent applies partial action by allocating only the necessary bit precision for actual accumulated values. This prevents excessive buffer allocation while maintaining sufficient precision for the specific application requirements
2Device complexity
If buffer capacity is reduced to decrease device complexity, then manufacturing cost and circuit scale are reduced, but measurement precision and reliability deteriorate due to overflow
Solution Approach 1:
The patent applies preliminary action by determining the appropriate bit precision before buffer allocation. By calculating the maximum accumulated value in advance and setting the bit precision accordingly, the system ensures reliability is maintained while avoiding excessive buffer capacity, thus resolving the contradiction between device complexity and reliability
Solution Approach 2:
The patent creates a simplified model of the accumulation process by calculating maximum values and determining required precision in advance, rather than allocating full capacity for all possible scenarios. This copying approach maintains reliability while reducing actual buffer requirements
3Measurement precision
If large buffer capacity is allocated to maintain processing accuracy, then measurement precision is improved, but productivity and processing speed are reduced due to increased processing load
Solution Approach 1:
The patent changes the parameter of bit precision from a large fixed value to a smaller dynamically determined value. This reduction in precision requirements directly decreases buffer capacity and processing load, thereby improving processing speed while maintaining sufficient accuracy for the application
4Device complexity
If bit precision is reduced to decrease buffer capacity, then device complexity is reduced, but measurement precision deteriorates due to information loss from truncation
Solution Approach 1:
The patent applies partial action by allocating exactly the necessary bit precision for the actual accumulated values rather than using excessive precision. By calculating the maximum accumulated value and determining the minimum required bits, the system achieves sufficient precision without unnecessary information representation, resolving the contradiction between device complexity and measurement precision
Data Source
AI summary
An information processing method comprises the steps of: inputting an input information of a multi-dimensional array; calculating an accumulated information value corresponding to a position of each element of the input information; and holding the accumulated information value in a buffer having a size of predetermined bits, wherein in the holding step, when an accumulated information value calculated in the calculating step overflows with respect to the size, a part not more than the predetermined bits of the calculated accumulated information value is held as the accumulated information value.