Speedup Robust Features based unsupervised place recognition for assistive mobile robot

Abstract

Vision Based qualitative localization or in the other word place recognition is an important perceptual problem at the center of several fundamental robot procedures. Place recognition approaches are utilized to solve the "global localization" problem. These methods are typically performed in a supervised mode. In this paper an appearance-based unsupervised place clustering and recognition algorithm are introduced. This method fuses several image features using Speedup Robust Features (SURF) by agglomerating them into the union form of features inside each place cluster. The number of place clusters can be extracted by investigating the SURF based scene similarity diagram between adjacent images. Experimental results show that this method is robust, accurate, efficient and able to create topological place clusters for solving the "global localization" problem with acceptable performance by the factor of clustering error and recognition precision.

Improving Street View Image Classification Using Pre-trained CNN Model Extracted Features

“…This problem is very challenging under active research, over the past few years, particularly with the increase in the number of images and datasets that are publicly available on the Internet, which provide an opportunity for research in predicting the geographic location of images. Due to the possibility of various applications in landmark recognition [1][2][3], urban reconstruction [4], place recognition [5][6][7], visual navigation [8,9], building recognition [10,11], and robot vision [12,13], attention from the research community has been drawn to the location estimation of images [14][15][16][17][18][19][20][21] over the past decade.…”

Two-dimensional laser-based environment exploration and recognition for service robots

“…Different from the above-mentioned methods, Cadena et al (2010) proposed a place recognition system by using stereo vision, which combines two key technologies: bag-of-word to detect loop closing candidate images, and conditional random fields to discard geometrical inconsistency. Karasfi et al (2011) developed an appearance-based place clustering and recognition method that fuses several image features by using speeded up robust features (SURF).…”