Infrared Image Recognition Method Based on Two-Stage Density Clustering

Abstract

The invention belongs to the technical field of image processing, and specifically relates to an infrared image recognition method based on two-stage density clustering, which includes two stages: using the Two-round-means algorithm to gather original data into blocks and form a representative point. The blocks are clustered using the improved CFDP algorithm. Eventually all nodes in each block get the same class labels as the representative points. The method provided by the invention greatly reduces the time complexity and space complexity of the algorithm, effectively improves the efficiency of the algorithm, and enables it to effectively cluster large-scale data sets; without any parameter setting, it is more efficient in actual use Concise, convenient, and better adaptability to various types of data sets.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to an infrared image recognition method based on two-stage density clustering. Background technique [0002] A large number of pictures collected by the far-infrared instrument are processed as experimental data, and the TSD clustering algorithm is used to analyze the data and judge the results. During analysis, the process of dividing a collection of physical or abstract objects into classes of similar objects is called clustering. A cluster generated by clustering is a collection of data objects that are similar to objects in the same cluster and different from objects in other clusters. [0003] Cluster analysis is based on similarity, with more similarities between patterns within a cluster than patterns not within the same cluster. [0004] Cluster analysis originates from many research fields, including data mining, statistics, machine learning, pattern recognition,...

AI Technical Summary

Problems solved by technology

[0024] 1. The algorithm has high time and space complexity, both O(n 2 ), not suitable for use in large datasets

[0025] 2. The algorithm is not very adaptable and can only be applied to special shape data sets

[0026] 3. The core parameters of the algorithm: the density threshold dc needs to be manually set

[0035] 1. The clustering performance of non-spherical data sets is not good, and data sets of arbitrary shapes cannot be effectively clustered

[0036] 2. Since the similarity between all samples and each class center is calculated every time, the convergence speed of the K-Means algorithm is relatively slow on large-scale data sets

Images