Vector-processor-oriented mean-residual normalized product correlation vectoring method

Abstract

The invention discloses a vector-processor-oriented mean-residual normalized product correlation vectoring method. The method comprises the following steps of: setting a reference graph A and a real-time graph B; traversing the real-time graph B and calculating a mean value of pixel values in the real-time graph B and an accumulated sum of pixel value squares Bij2 respectively; traversing the reference graph A and taking two sub graphs Auv and A(u+4)v from the reference graph A each time, and shuffling to obtain four sub graphs A(u+k)v (k=0, 1, 2 and 3); sequentially calculating the accumulated sum of the pixel values, the accumulated sum of (A(u+k)v)ij2 and the accumulated sum of (A(u+k)v)ij*Bij; sequentially calculating the mean-residual normalized product correlation coefficients of the sub graphs A(u+k)v (k=0, 1, 2 and 3) with the real-time graph B; and setting u to be u+4, repeating the steps until the reference graph A is traversed completely so as to acquire all the mean-residual normalized product correlation coefficient values.

Description

technical field [0001] The invention relates to the field of image matching and vectorized compilation thereof, in particular to a vectorized realization method of mean-removed normalized product correlation coefficients. Background technique [0002] With the increasingly high computing requirements of computing-intensive applications such as 4G wireless communication, radar signal processing, high-definition video and digital image processing, it is difficult for a single chip to meet the application requirements. Multi-core processors, especially vector processors, have been widely used application. Vector processors generally consist of multiple processor elements (PEs) and typically support vector-based data loads and stores. Each PE contains multiple independent functional units, generally including shift units, ALU units, multiplication units, etc. Vector processors usually support SIMD (Single Instruction / Multiple Data) operations, that is, under the control of the...

AI Technical Summary

Problems solved by technology

But for a vector processor, on the one hand, the vector processor contains multiple processor units, and it is difficult to reuse the previous calculation results. On the other hand, the sub-image pixel data usually uses 8-bit pixel values, and it needs to be byte-by-byte when traversing the reference image. Read image data with an offset, and vector processors generally do not support data reading across word boundaries

At present, there is a lack of effective vectorization implementation methods for vector processor-oriented normalized normalized product correlation coefficients

Images