Online Learning With Novelty Detection in Human-Guided Road Tracking

Abstract

Abstract-Current image processing and pattern recognition algorithms are not robust enough to make automated remote sensing image interpretation feasible. For this reason, we need to develop image interpretation systems that rely on human guidance. In this paper, we tackle the problem of semi-automatic road tracking in aerial photos. We propose an online learning approach that naturally integrates inputs from human experts with computational algorithms to learn road tracking. Human inputs provide the online learner with training examples to generate road predictors. An ensemble of road predictors is learned incrementally and used to automatically track roads. When novel situations are encountered, control is returned back to the human expert to initialize a new training and tracking iteration. Our approach is computationally efficient, and it can rapidly adapt to dynamic situations where the image feature distributions change. Experimental results confirm that our approach is effective and superior to existing methods.

Road Feature Extraction from High Resolution Aerial Images Upon Rural Regions Based on Multi-Resolution Image Analysis and Gabor Filters

“…To qualitatively evaluate the performance of the semi-automatic road extraction algorithms, four criteria (correctness, completeness, efficiency, and accuracy) are utilized in (Zhou et al, 2006) and further in (Zhou et al, 2007). Completeness and correctness are the priority criteria in cartography, while the efficiency measurement principally takes the savings of human input into consideration.…”

Delineation and geometric modeling of road networks

“…The approach was successfully applied on rural as well as semi-urban areas with successful results. Zhou et al (2007) present a user-guided image interpretation system which integrates inputs from human experts with computational algorithms in order to learn road tracking. Although the results seem promising, the goal of completely eliminating the need for human intervention and interactions is still not achieved.…”

Bayesian Tracking of Linear Structures in Aerial Images

“…In contrast, we use dynamic tracking because the system is imbedded in a semi-automatic system for interpreting aerial images. In this system, the computer acts as an assistant for the human operator, learns simple image interpretations online, takes over simple tasks such as tracking roads or pipelines, and returns control to the operator whenever a problem is encountered [14,15]. It is thus important that the computer proceeds in a sequential manner, permitting the operator to stop the interpretation process at any moment without having to correct or delete too many incorrect interpretations.…”

“…Several strategies have been pursued to overcome this problem. First, one can design of semi-automatic systems, where the computer acts as an assistant to the human operator, taking over very simple tasks, and returning control to the operator whenever a problem is encountered [14,15]. Second, one can focus on systems composed of many modules, each specialized on one particular image interpretation task.…”