Clustering Algorithms: Taxonomy, Comparison, and Empirical Analysis in 2D Datasets

Abstract

Because of the abundance of clustering methods, comparing between methods and determining which method is proper for a given dataset is crucial. Especially, the availability of huge experimental datasets and transactional and the emerging requirements for data mining and the like needs badly for clustering algorithms that can be applied in various domains. This paper presents essential notions of clustering and offers an overview of the significant features of the most common representative clustering algorithms of clustering categories presented in a comparative way. More specifically the study is based on the numerical type of the data that the algorithm supports, the shape of the clusters, and complexity. The experiments were done using nine clustering algorithms representing the common clustering categories on eight 2D clustered datasets differ in the clusters' shapes and density of the data points. Furthermore, the comparison was done from the point of view seven performance measures.

Monocular Visual SLAM for Markerless Tracking Algorithm to Augmented Reality

“…The feature-based SLAM extracts feature pairs from input image feature points and descriptors, and calculates camera poses and maps by matching 2D to 2D, 2D to 3D, and 3D to 3D feature points. Visual SLAM is separated into filter-based SLAM methods and nonlinear optimization-based SLAM methods according to the method of back-end optimization [9]. The PTAM (Parallel tracking And Mapping) [10] is a typical monocular visual SLAM system based on key-frames and using nonlinear optimization.…”

“…1. When the plane is not found in the SLAM system, the initial pose of the model is set to T 0 , and the SLAM system transmits the current frame pose and the current frame image data to the AR module; when the plane is found in the SLAM system, the pose of the model is set to T pw , and the SLAM system still transmits the current frame pose and the current frame image data to the AR module [17][18][19]. A buffered Blocking queue severs as the transmission container between the SLAM system and the augmented reality module.…”