Entity Association Using Context for Wide Area Motion Imagery Target Tracking
| Autor: | Peiyi Li, Erik Blasch, Xinchu Shi, Pengpeng Liang, Haibin Ling |
|---|---|
| Rok vydání: | 2016 |
| Předmět: | |
| Zdroj: | Context-Enhanced Information Fusion ISBN: 9783319289694 Context-Enhanced Information Fusion |
| Popis: | Entity estimation includes tracking the numbers and types of targets in a scene and is challenging in large area surveillance due to high target density, severe similar target ambiguity, and a low sensor frame rate. Moving vehicle detection from wide area aerial surveillance can be aided by context information. In this paper, we utilize the maximum consistency context (MCC) as spatiotemporal information to estimate multiple targets, and temporal context (TC) to capture the road information. For a candidate association, the MCC is defined as the most consistent association in its neighborhood. Such a maximum selection chooses the reliable neighborhood context information while filtering out noisy and distracting data. In contrast with previous methods to exploit road information, TC does not need to get the location of the road first or use the geographical information systems’ (GIS) information. We first use background subtraction to generate the candidates and then build MCC/TC based on the candidates that have been classified as positive by histograms of oriented gradient (HOG) with multiple kernel learning (MKL). For each positive candidate, a region around the candidate is divided into several subregions based on the direction of the candidate, then each subregion is divided into 12 bins with a fixed length; and finally the TC, a histogram, is built according to the positions of the positive candidates in eight consecutive frames. In order to benefit from both the appearance and context information, we use MKL to combine MCC/TC and HOG. We demonstrate the usefulness of context modeling on multi-target tracking using three challenging wide area motion imagery (WAMI) sequences using the publicly available Columbus Large Image Format (CLIF) 2006 dataset. Both quantitative and qualitative results show clearly the effectiveness of using MCC and TC information, in comparison with algorithms that use no context information. Likewise, the experiments demonstrate that the proposed MCC/TC are useful to remove the false positives that are away from the road and the combination of TC and HOG with MKL outperforms the use of TC or HOG only. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|