Sparse matching

Name: “sparse_matching”

Description

Compute keypoints matches on pair images

Architecture Note

This application uses a plugin-based architecture separating the application layer from the method layer:

• Application Parameters: These control the parallelization at the application level (tiling, margins, match validation, etc.) and are independent of the matching algorithm used.

• Method Parameters: These are algorithm-specific parameters (e.g., SIFT tuning parameters) and belong to the selected method plugin.

The application parameter selects which parallelization strategy to use. The method parameter selects which matching algorithm/preset to use.

Both application and method may have their own parameters, which should then be put all together in the sparse_matching configuration key.

Applications

Basic Application

Name	Description	Type	Available value	Default value	Required
application	Application to use in the pipeline	string	“basic”	“basic”	No
method	Method to use for sparse matching	string	“sift”, “pandora2d”	“sift”	No
elevation_delta_lower_bound	Expected lower bound for elevation delta with respect to input low resolution dem in meters	int, float		None	No
elevation_delta_upper_bound	Expected upper bound for elevation delta with respect to input low resolution dem in meters	int, float		None	No
epipolar_error_upper_bound	Expected upper bound for epipolar error in pixels	float, str	should be > 0	auto	No
epipolar_error_estimation	Mean for epipolar error in pixels	float, str	should be >= 0	auto	No
epipolar_error_maximum_bias	Maximum bias for epipolar error in pixels	float, str	should be >= 0	auto	No
save_intermediate_data	Save matches in epipolar geometry (4 first columns) and sensor geometry (4 last columns)	boolean		false	No
tile_margin	Margin to use on tiles	int	should be > 0	10	No
minimum_nb_matches	Minimum number of matches that must be computed to continue pipeline	int	should be > 0	90	No
decimation_factor	Reduce the number of keypoints	int	should be > 0	30	No
disparity_bounds_estimation	Parameters for the estimation of disparity interval	dict		{}	No

Note

For the decimation factor, a value of 33 means that we divide the number of sift by 3, a value of 100 means that we do not decimate them

Note

‘elevation_delta_lower_bound’ and ‘elevation_delta_upper_bound’ are overidden to [-1000, 9000] in default pipeline if no initial elevation is set. If initial elevation is set, it is overridden to [-500, 1000].

Disparity bounds estimation:

Name	Description	Type	Available value	Default value	Required
activated	activates estimation of disparity interval from SIFT matches	bool		True	No
percentile	percentage of SIFT matches to ignore	int		1	No
upper_margin	margin (in meters) added to altitude of higher SIFT match retained	int		1000	No
lower_margin	margin (in meters) substracted from altitude of lower SIFT match retained	int		500	No

Methods

SIFT Method

Name: “sift”

Description: Scale-Invariant Feature Transform (SIFT) based sparse matching

The SIFT method parameters are automatically selected from the sparse matching configuration above when method is set to “sift”.

Method-specific Parameters:

Name	Description	Type	Available value	Default value	Required
sift_matching_threshold	Threshold for the ratio to nearest second match	float	should be > 0	0.7	No
sift_n_octave	The number of octaves of the Difference of Gaussians scale space	int	should be > 0	8	No
sift_n_scale_per_octave	The numbers of levels per octave of the Difference of Gaussians scale space	int	should be > 0	3	No
sift_peak_threshold	Constrast threshold to discard a match (at None it will be set according to image type)	float	should be > 0	4.0	No
sift_edge_threshold	Distance to image edge threshold to discard a match	float		10.0	No
sift_magnification	The descriptor magnification factor	float	should be > 0	7.0	No
sift_window_size	smaller values let the center of the descriptor count more	int	should be > 0	2	No
sift_back_matching	Also check that right vs. left gives same match	boolean		true	No
used_band	Name of band used for correlation	int	should be in input sensor	“b0”	No
tile_width	the tile width	int		5000	No
tile_height	the tile height	int		60	No

For more information about SIFT parameters, please refer to the VLFEAT SIFT documentation.

Pandora2d Method

Name: “pandora2d”

Description: Pandora2D is an Image Registration framework. It computes 2D displacement maps from a pair of images taken over the same scene. It uses Pandora.

The PANDORA2D method parameters are automatically selected from the sparse matching configuration above when method is set to “pandora2d”.

Method-specific Parameters:

Name	Description	Type	Available value	Default value	Required
step	The step defined for matching cost	list	should be higher than 0	[10, 10]	No
conf_to_use	The conf to use in pandora2d	str		“default”	No
used_band	Band used for correlation	str	should be in input sensor	“b0”	No
loader	External library used to compute sparse matching	str	“pandora”	“pandora”	No
loader_conf	Configuration for Pandora2d loader (for pandora_custom method)	dict or str		None	No
threshold_disp_range_to_borders	Clip the disparity range to the valid region of right image	bool		False	No
tile_width	the tile width	int		500	No
tile_height	the tile height	int		60	No

Examples

Minimal example:

YAML JSON Copy

---
applications:
  sparse_matching:
    application: basic
    method: sift

{
    "applications": {
        "sparse_matching": {
            "application": "basic",
            "method": "sift"
        }
    }
}

Example with both application and method parameters:

YAML JSON Copy

---
applications:
  sparse_matching:
    application: basic
    method: sift
    tile_margin: 12
    minimum_nb_matches: 120
    decimation_factor: 40
    disparity_bounds_estimation:
      activated: true
      percentile: 1
      lower_margin: 500
      upper_margin: 1000
    sift_matching_threshold: 0.7
    sift_n_octave: 8
    sift_peak_threshold: 4.0
    sift_back_matching: true
    used_band: b0

{
    "applications": {
        "sparse_matching": {
            "application": "basic",
            "method": "sift",
            "tile_margin": 12,
            "minimum_nb_matches": 120,
            "decimation_factor": 40,
            "disparity_bounds_estimation": {
                "activated": true,
                "percentile": 1,
                "lower_margin": 500,
                "upper_margin": 1000
            },
            "sift_matching_threshold": 0.7,
            "sift_n_octave": 8,
            "sift_peak_threshold": 4.0,
            "sift_back_matching": true,
            "used_band": "b0"
        }
    }
}