Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Graphic processing method based on logarithm segmentation linear approximation

A piecewise linear approximation, graphics processing technology, applied in complex mathematical operations and other directions, can solve the problem of low precision and achieve the effect of simplifying complex operations

Inactive Publication Date: 2013-08-28
SHANGHAI UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The main purpose of the present invention is to provide a graphic processing method based on logarithmic piecewise linear approximation for the defects of low precision existing in the prior art, based on error analysis and control of logarithmic piecewise linear approximation, by setting the corresponding maximum relative error To obtain a specific segmented lookup table LUT, thereby improving the graphic processing method of logarithmic conversion calculation accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Graphic processing method based on logarithm segmentation linear approximation
  • Graphic processing method based on logarithm segmentation linear approximation
  • Graphic processing method based on logarithm segmentation linear approximation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Embodiment 1: This embodiment is based on the graph processing method of logarithmic piecewise linear approximation, and the operation steps of this processing method are as follows:

[0032] (1) According to the principle of logarithmic piecewise linear approximation, find out x i and x i+1 , , x i+1 are the first and last points of each segmented interval respectively, and then calculate a i and d i , a i Indicates the control parameter of the degree of approximation of each interval piecewise straight line, d i Represents the intercept, constituting the look-up table LUT.

[0033] (2) Will x As input to the lookup table, x Operators for multiplication, division and exponent operations in graphics processing systems, x and the endpoints of each interval , x i+1 compare to find x The interval where it is located, find out the 4 parameters corresponding to the interval p, q, r, d , p, q, r express x the number of shifts to perform, d Indicates t...

Embodiment 2

[0036] Embodiment 2: This embodiment is basically the same as Embodiment 1, and its special features are as follows: the specific steps of forming the lookup table LUT in the step (1) are as follows:

[0037] (1-1) Set the maximum relative error of two pre-set parameters E Rmax and the first endpoint of the first segment interval x 1 =1.0;

[0038] (1-2) According to the parameters in step (1-1) x 1 , E Rmax and the following formula, it can be calculated that x 2 ;

[0039]

[0040]

[0041] In the formula x 1 、x 2 is the endpoint of the interval, x’ is the value point of the maximum relative error, E R is the relative error;

[0042] (1-3) will x 2 As the first endpoint of the second segment interval, and then repeat steps (1-2), you can get x 3 , and then iterate until the last interval endpoint is found x i ;

[0043] (1-4) According to all the segmentation points obtained in step (1-3), according to the following formula, calculate ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a graphic processing method based on logarithm segmentation linear approximation. The method comprises the following steps: according to a logarithm segmentation linear approximation theory, finding out xi and xi+1 which are respectively a head point and a tail point of each segmentation region; then calculating ai and di to form a lookup table LUT, wherein the ai represents a control parameter of the segmentation linear approximation degree of each region; comparing x with the end points of each region and the xi+1 by taking the x as an input of the lookup table to find out the region where the x is positioned, wherein the x is operators of multiplication, division and index calculation in a graphic processing system, and looking up to four parameters, namely p, q, r and d of the corresponding region, wherein p, q and r represents shifting bits performed on the x; according to the obtained parameters, shifting the input parameter x, and respectively shifting the x by p, q and r, wherein obtained shifting results are respectively represented as x1', x2' and x3'; and summing the x1', x2' and x3' into d, and obtaining a summed result s. By the method for instructing logarithm segmentation by relative error control, the error of logarithm transformation can be reduced, and the calculation precision of the graphic processing method is greatly improved.

Description

technical field [0001] The invention relates to a graph processing method based on logarithm segmental linear approximation, which is a graph processing method using error analysis to guide logarithm segmental linear approximation. Background technique [0002] With the popularization of mobile smart terminals, graphics processing capabilities have received more and more attention. In embedded graphics processing systems, graphics processing requires a large number of matrix multiplication, vector unitization, vector multiplication, exponential operations and averaging. Complex calculations such as square root, and smart terminals have the characteristics of battery power supply, so there are very large restrictions on the power consumption and area of ​​graphics processing chips. Therefore, directly using the general number system to calculate relatively complex operations in graphics processing will increase the chip. Power consumption and increase the area, in order to si...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F17/10
Inventor 朱梦尧高留闯郑佳贺君炜万旺根
Owner SHANGHAI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products