Shading Correction Device Using Bit-Shift Lookup Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing shading correction devices face high processing costs and long processing times due to division calculations in image data, leading to large lookup table sizes when handling images with numerous pixel values.
Innovation Solution
A shading correction device that reduces data size by using a lookup table storing bit-shifted reference pixel values, allowing for efficient calculation of shading correction values through bit shift and multiplication operations, with a second lookup table for storing bit shift values, thereby minimizing the overall data size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If division calculation is used for shading correction, then calculation accuracy is maintained, but processing time increases and processing cost increases
Solution Approach 1:
The patent pre-calculates shading correction values for all possible reference pixel value ranges and stores them in lookup tables (LUTs) before actual image processing. During shading correction, the system only needs to perform a simple table lookup and bit shift operation instead of real-time division calculation, thus maintaining accuracy while dramatically reducing processing time.
Solution Approach 2:
The patent replaces the complex division calculation mechanism with a simpler lookup table query mechanism combined with bit shift operations. By substituting the mathematical division operation with pre-computed table lookups and bitwise operations, the system achieves the same shading correction effect with significantly lower computational cost.
2Measurement precision
If lookup table stores all reference pixel values, then shading correction accuracy is maintained, but data size of lookup table becomes large
Solution Approach 1:
The patent divides the reference pixel value range into multiple segments or groups, creating separate lookup tables for different value ranges. This segmentation allows the system to store pre-calculated correction values for each segment, reducing the memory access complexity and enabling more efficient table lookups while maintaining comprehensive coverage of all possible pixel values.
Solution Approach 2:
The patent introduces bit shift operations as an additional dimension to the lookup table approach. By combining table lookup with dynamic bit shifting based on the reference pixel value, the system can efficiently handle a wide range of pixel values without needing to store separate tables for every possible value, thus reducing overall data size while maintaining accuracy.
Data Source
AI summary
A shading correction device includes a reference acquirer that acquires a reference pixel value W in a captured image of a color measurement reference object, a target pixel value acquirer that acquires a pixel value S in a captured image of a measurement target, a bit shift value acquirer that acquires a bit shift value k2 based on the reference pixel value W, a lookup table storing a variable G2 (=2k1/W>>k2) corresponding to W>>k2, and a calculator that reads the variable G2 corresponding to the reference pixel value W and the bit shift value k2 from the lookup table and calculates a shading correction value Rdec of each pixel in the captured image of the measurement target using Expression (5):Rdec=(F·G2)⪢(k1+k2)+BwhereF=S·(A-B)+{W⪢1},G2=2k1W⪢k2,(5)and A, B and k1 are integer constants.