Map2DFusion: Real-time Incremental UAV Image Mosaicing based on Monocular SLAM

We present a real-time approach to stitch large-scale aerial images incrementally. A monocular SLAM system is used to estimate camera position and attitude, and meanwhile 3D point cloud map is generated. When GPS information is available, the estimated trajectory is transformed to WGS84 coordinates after time synchronized automatically. Therefore, the output orthoimage retains global coordinates without ground control points. The final image is fused and visualized instantaneously with a proposed adaptive weighted multiband algorithm. To evaluate the effectiveness of the proposed method, we create a publicly available aerial image dataset with sequences of different environments. The experimental results demonstrate that our system is able to achieve high efficiency and quality compared to state-of-the-art methods. Detailed information can be found at: http://zhaoyong.adv-ci.com/map2dfusion

Source code of Map2DFusion can be downloaded at: https://github.com/zdzhaoyong/Map2DFusion

1. Demonstration

Map2DFusion

Online video is available at: https://www.youtube.com/watch?v=-kSTDvGZ-YQ

The video is composed of three parts:

  • Part 1: Real-time reconstruction with GPS (Sequence phantom3-village)
  • Part 2: Experiment on Challenging Environment (Sequence phantom3-grass)
  • Part 3: Ground Control Station Intergreted Map2DFusion (Sequence gopro-npu)

2. Results on NPU DroneMap Dataset

An evaluation of our system is performed over the NPU DroneMap dataset, where thumbnails of the results are shown in the figure below.

Map2DFusion Results

We compare orthoimages generated by our method with two well-known softwares Photoscan and Pix4DMapper. For software Photoscan and Pix4DMapper, the parameter for aligning images is set to ‘precise level’. The full resolution results are available in the table below and images can be downloaded by either click the sequence or the thumbnails.

3. Downloads

Comments are closed.