Abstract
This study was conducted to develop original benchmark datasets that simultaneously include indoor–outdoor visual features. Indoor visual information related to images includes outdoor features to a degree that varies extremely by time, weather, and season. We obtained time-series scene images using a wide field of view (FOV) camera mounted on a mobile robot moving along a 392-m route in an indoor environment surrounded by transparent glass walls and windows for two directions in three seasons. For this study, we propose a unified method for extracting, characterizing, and recognizing visual landmarks that are robust to human occlusion in a real environment in which robots coexist with people. Using our method, we conducted an evaluation experiment to recognize scenes divided up to 64 zones with fixed intervals. The experimentally obtained results using the datasets revealed the performance and characteristics of meta-parameter optimization, mapping characteristics to category maps, and recognition accuracy. Moreover, we visualized similarities between scene images using category maps. We also identified cluster boundaries obtained from mapping weights.
Highlights
With the rapid progress of recent artificial intelligence (AI) and robotic technologies [1], widely various intelligent robots [2] have been developed for industrial utilization at factories and warehouses
For our earlier study [32], we proposed a unified method for extracting, characterizing, and recognizing visual landmarks (VLs) that were robust to human occlusion in a real environment in which robots coexist with people
histograms of oriented gradients (HOGs) [35] was used for our earlier study [32], the human detection accuracy with HOG, which includes insufficient robustness to rotation and scaling, was increased by only 60%
Summary
With the rapid progress of recent artificial intelligence (AI) and robotic technologies [1], widely various intelligent robots [2] have been developed for industrial utilization at factories and warehouses. They have been developed for collaborative utilization in human societies and facilities in terms of homes [3], offices [4], kindergartens [5], nursing-care facilities [6], and hospitals [7]. Robots must have capabilities to perform accurate actions and functions of self-localization, path planning and tracking, object recognition, and environmental understanding [8]. Simultaneous localization and mapping (SLAM) technologies [10] have been studied widely as a fundamental approach for autonomous locomotion of mobile robots including drones [11].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
