Image pyramid feature extraction circuit, device and method

Abstract

The invention relates to an image pyramid feature extraction circuit. A parallel reusable read-in circuit reads a base image frame, calculates the coordinates of N valid data lines of the M-th layer of an image pyramid in the base image frame, and applies and stores N valid data lines according to the coordinates. An image pyramid feature calculation circuit comprises M image pyramid feature extraction units, and simultaneously carries out feature calculation of the M-th layer of the image pyramid according to N valid data lines. An image pyramid feature converging output circuit converges and outputs M*K image pyramid features output by the image pyramid feature calculation circuit to a memory. M and N are positive integers greater than or equal to 2, and K is the number of feature channels of the image pyramid feature calculation unit. Hardware is used to realize image pyramid feature extraction without carrying out more preprocessing and reducing the accuracy algorithm optimization. The image pyramid feature extraction circuit has the advantages of higher calculation speed and lower bandwidth. The invention further provides an image pyramid feature extraction device and method.

Description

technical field [0001] The invention relates to the fields of image signal processing and integrated circuit design, in particular to an image pyramid feature extraction circuit, device and method. Background technique [0002] Decomposing an image with multiple resolutions and multiple angles is a basic image analysis method. Obviously, when the method of representing the image is over-complete, it will be easier to extract the required visual information from it, and the accuracy of the information obtained by calculation is more accurate at this time. For example, it is a very effective method to construct an image feature pyramid as an information source for target recognition, and to improve the recognition accuracy through step-by-step calculations. [0003] Feature extraction from image pyramids is widely used in machine vision algorithms. Such as SIFT feature extraction, Ada-Boost detection, Lucas-Kanade optical flow method, etc. In this kind of machine vision alg...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at the above defects of the prior art, and provides an image pyramid feature extraction circuit, which overcomes the defects of large amount of calculation and serious bandwidth consumption of the traditional image pyramid feature extraction algorithm based on software processing

Images