V. S. Sidorova

  • Citations Per Year
Learn More
A histogram clustering algorithm is suggested, which builds the hierarchy of distributions better in cluster separability. The algorithm optimizes the average cluster separability choosing the system of the data subdomain quantization grid and allows a significant decrease in the number of clusters. Application of the algorithm for uncontrolled Earth’s(More)
In the capacity of texture features, the parameters of the SAR model for unlabeled classification image of forest landscape are used. We compare the results for two systems of features: the SAR model and the popular Haralik statistics. The histogram clustering algorithm arrange the hierarchy of distributions better by cluster isolation.
A new automatic, hierarchical, multidimensional, histogram-based clusterization algorithm is considered. A method for choosing the clusterization detailedness in different regions of the vector space of spectral features depending on the average separability of clusters is proposed. The algorithm is applied for the automatic classification of multispectral(More)
The proposed histogram-based algorithm searches for the clustering detailedness that differs in subdomains of the vector space of spectral features depending on the average separability of clusters. The objective of the hierarchical decomposition of clusters is to achieve limit detailedness with respect to the given cluster separability. Application of the(More)
Automatic clusterization and follow segmentation of aerial pictures images by textural features is are considered. A divisible histogram hierarchical algorithm with search of clusters with preset separability is used. The peculiarity of segmentation by statistical textural features are taken into account. The parameters of the model for SAR imaging are used(More)
  • 1