SIFT algorithm key point detection method based on hardware circuit

Abstract

The invention discloses an SIFT (Scale Invariant Feature Transform) algorithm key point detection method based on a hardware circuit, which comprises the following steps of: unifying image layers andpixel points thereof required for constructing a differential Gaussian scale space in a key point detection space in a mode of acquiring a multi-scale image layer and multi-point image pixels at one time by a register; determining accurate key points by extreme point detection and two times of extreme point accurate positioning extreme points so that complexity of obtaining the regional information of the to-be-detected point is greatly reduced. Compared with the prior art, the technical scheme has the advantages that the data acquisition and analysis time in the extreme point detection process is effectively reduced, and the hardware flow complexity is reduced, so that the algorithm efficiency is improved, the hardware circuit area is saved, and the SIFT algorithm key point detection real-time performance is improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a hardware circuit-based SIFT algorithm key point detection method. Background technique [0002] SIFT, or Scale-invariant feature transform (SIFT), is a description used in the field of image processing. This description has scale invariance, can detect key points in the image, and is a local feature descriptor. The SIFT feature is based on some local appearance interest points on the object independent of the size and rotation of the image. The tolerance to light, noise, and micro viewing angle changes is also quite high. Widely used in video tracking, image 3D modeling, object recognition, image panorama stitching and other fields. [0003] Due to the large amount of calculation of the SIFT algorithm, in practical applications, with the improvement of the performance of the camera, the resolution of the image is getting higher and higher, the amount of information ...

AI Technical Summary

Problems solved by technology

[0004] In the process of executing the SIFT algorithm, the pixel information of the differential Gaussian scale space is stored in the SRAM inside the chip. In order to obtain more accurate extreme points, it is necessary to change the position of the interpolation center repeatedly. After changing the interpolation center, it is necessary to read the SRAM Address management can become very complicated

Under the existing logic control conditions for reading and writing SRAM addresses, multiple data collection and analysis will increase the complexity of hardware design and reduce the efficiency of the algorithm, which will have a certain impact on the real-time performance of the algorithm

Images