Feature Description
Connection points refer to corresponding image points that can construct stereo models or establish connection relationships between adjacent models (images). Generate Connection Points is the first core step in the UAV data processing workflow, providing high-precision observations for subsequent aerial triangulation.
This function adopts the SIFT (Scale-Invariant Feature Transform) algorithm to extract local features that are robust to rotation, scaling, and illumination changes from images, and establishes reliable corresponding point relationships under different viewpoints through feature descriptor matching. The generated connection points will serve as input for aerial triangulation adjustment, ensuring the consistency of images in spatial position and fundamentally affecting the geometric accuracy of products such as DOM and DSM.
In the following scenarios, you may use this function:
- Before processing large-scale UAV oblique or orthophoto imagery tasks, it is necessary to generate high-quality connection points in bulk.
- When images have weak texture, repeated texture, or large viewpoint changes, adjust the number of seed points to improve the matching success rate.
- In scenarios requiring quick preview or low precision, enabling downsampling can significantly reduce the time for feature extraction and matching.
This function is only used for the Generate Connection Points operation in the UAV data processing workflow.
SuperMap ImageX Pro 2026 版本开始支持。
Function Entry
UAV Data Processing Workflow tab ->Imagery group ->Generate Connection Points.
Parameter description
| Parameter Name | parameter interpretation | parameter type |
| Matching Method | Specifies the matching algorithm between images. The current version only provides the SIFT method, which is a method for extracting unique invariant features from images and can be used for reliable matching between objects or scenes under different viewpoints. | ImageMatchMethod |
| Number of Seed Points | Sets the maximum number of candidate seed points during the matching process. The larger the value, the longer the matching time, but it can increase the number of connection points in weak texture areas, thereby improving the robustness of aerial triangulation solution. The default value is 4096. | Integer |
| Downsampling Factor | Performs downsampling on the original image, which can improve the speed of feature extraction and matching, but may lose partial match accuracy. The default value is 1, meaning no downsampling. If the value is greater than 1, downsampling is performed. | Integer |
Output Result
Outputs the connection point information for each image. It needs to be viewed in Correct Camera Points after the aerial triangulation operation.
Related Topics