Target identification method based on image characteristic analysis

Abstract

The invention relates to a target identification method based on image characteristic analysis, which comprises the following steps: selecting 2*2 subregions by using a scale invariant feature transform (SIFT) characteristic point as the center; normalizing the gradient amplitudes of each subregion in 8 gradient directions, selecting the gradient direction with the maximum gradient amplitude from the 8 gradient directions, rotating the gradient direction with the maximum gradient amplitude to X-axis or Y-axis, and selecting 4 gradient directions positioned on X-axis and Y-axis; and computing and comparing the magnitude of the gradient amplitudes, and selecting the characteristic point with centralized gradient directions as the SIFT characteristic point of the image within the range of the characteristic point vector matching thresholds. The method can be used to reduce the number of dimensions of the SIFT characteristic vector and the quantity of the characteristic points without influencing the image matching effect, raise the identification speed and accuracy for the environmental target, and enhance the real-time on-line running speed and matching accuracy of the entire algorithm.

Description

technical field [0001] The invention relates to a target recognition method based on image feature analysis, in particular to a target recognition method based on image feature analysis in the local environment field where an intelligent mobile robot detects an unknown space. Background technique [0002] More than 70% of human beings' acquisition of external environment information comes from vision. It can be seen that visual navigation is a main development direction of intelligent mobile robot navigation. The literature published in the prior art (Li Guizhi and An Chengwan. Research on the positioning method of mobile robots based on scene recognition [J]. Robotics, 2005.27 (2): 23-26) proposed a charge-coupled device based , CCD) video acquisition to identify the mobile robot positioning method for the scene, the mobile robot is positioned by the global feature scene recognition, the global texture feature of the image is extracted through the Gabor filter, and it is ca...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the robot collects a lot of scene sample images in an unknown environment, and the vector machine, which belongs to the small sample recognition problem, is not suitable for the large sample scene matching problem based on global features.

[0004] However, in the above SIFT method, the standard SIFT feature description vector dimension has 128 bits, which requires a lot of resources in the calculation and storage process

Moreover, there are many feature points in each image. For example, the number of feature points in a 150-image library can reach nearly 40,000. Such a huge amount of data not only imposes an important burden on real-time online feature matching calculations, but also easily forms wrong matches.

Images