Object Detection and Recognition by Template Matching Using SIFT Features for Scene Image of Room

Assistance of Invisible Floor Marks, Partial Templates, Spatial Relation to Floor, and Information on Object Shape Detected by Haar-Like Features

Abstract

For the purpose of simultaneous localization and mapping (SLAM) of an indoor mobile robot, a method to process only one monocular image (640x480) of the whole view of environment is proposed. This method imitates that a human can grasp momentarily the situation of a room, i.e., the layout of objects/obstacles. It is not required for a robot to move around the room while probing the objects/obstacles by some sensors, which is usually done in SLAM research. Object recognition is done by using scale invariant feature transform (SIFT), partial template matching, and object relative restriction. Additionally, verification of object category by AdaBoost classifier using Haar-like features based on information on the object shape.