pitsurf 0.0.0a1__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- pitsurf-0.0.0a1/LICENSE +21 -0
- pitsurf-0.0.0a1/PKG-INFO +131 -0
- pitsurf-0.0.0a1/README.md +92 -0
- pitsurf-0.0.0a1/pitsurf/__init__.py +6 -0
- pitsurf-0.0.0a1/pitsurf/core/__init__.py +0 -0
- pitsurf-0.0.0a1/pitsurf/core/interpolation.py +160 -0
- pitsurf-0.0.0a1/pitsurf.egg-info/PKG-INFO +131 -0
- pitsurf-0.0.0a1/pitsurf.egg-info/SOURCES.txt +13 -0
- pitsurf-0.0.0a1/pitsurf.egg-info/dependency_links.txt +1 -0
- pitsurf-0.0.0a1/pitsurf.egg-info/requires.txt +25 -0
- pitsurf-0.0.0a1/pitsurf.egg-info/top_level.txt +1 -0
- pitsurf-0.0.0a1/pyproject.toml +41 -0
- pitsurf-0.0.0a1/setup.cfg +4 -0
- pitsurf-0.0.0a1/tests/test_nearest_region.py +19 -0
- pitsurf-0.0.0a1/tests/test_sector_idw.py +55 -0
pitsurf-0.0.0a1/LICENSE
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026 Leonid Shpaner
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
pitsurf-0.0.0a1/PKG-INFO
ADDED
|
@@ -0,0 +1,131 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: pitsurf
|
|
3
|
+
Version: 0.0.0a1
|
|
4
|
+
Summary: Surface construction from scattered survey points: polygonal (nearest-region), B-spline, NURBS, and SVR interpolators for open-pit mine surveying.
|
|
5
|
+
Author: Leonid Shpaner
|
|
6
|
+
License: MIT
|
|
7
|
+
Project-URL: Homepage, https://github.com/lshpaner/pitsurf
|
|
8
|
+
Project-URL: Repository, https://github.com/lshpaner/pitsurf
|
|
9
|
+
Keywords: mine surveying,interpolation,open-pit,NURBS,IDW,surface modeling
|
|
10
|
+
Classifier: Development Status :: 2 - Pre-Alpha
|
|
11
|
+
Classifier: Intended Audience :: Science/Research
|
|
12
|
+
Classifier: License :: OSI Approved :: MIT License
|
|
13
|
+
Classifier: Programming Language :: Python :: 3
|
|
14
|
+
Classifier: Topic :: Scientific/Engineering
|
|
15
|
+
Requires-Python: >=3.10
|
|
16
|
+
Description-Content-Type: text/markdown
|
|
17
|
+
License-File: LICENSE
|
|
18
|
+
Requires-Dist: numpy>=1.24
|
|
19
|
+
Requires-Dist: scipy>=1.11
|
|
20
|
+
Provides-Extra: kernel
|
|
21
|
+
Requires-Dist: scikit-learn>=1.3; extra == "kernel"
|
|
22
|
+
Provides-Extra: nurbs
|
|
23
|
+
Requires-Dist: geomdl>=5.3; extra == "nurbs"
|
|
24
|
+
Provides-Extra: data
|
|
25
|
+
Requires-Dist: pandas>=2.0; extra == "data"
|
|
26
|
+
Requires-Dist: pyarrow>=14; extra == "data"
|
|
27
|
+
Requires-Dist: rasterio>=1.3; extra == "data"
|
|
28
|
+
Requires-Dist: typer>=0.9; extra == "data"
|
|
29
|
+
Provides-Extra: viz
|
|
30
|
+
Requires-Dist: matplotlib>=3.7; extra == "viz"
|
|
31
|
+
Provides-Extra: dev
|
|
32
|
+
Requires-Dist: pytest; extra == "dev"
|
|
33
|
+
Requires-Dist: black; extra == "dev"
|
|
34
|
+
Requires-Dist: nbstripout; extra == "dev"
|
|
35
|
+
Requires-Dist: jupyter; extra == "dev"
|
|
36
|
+
Requires-Dist: build; extra == "dev"
|
|
37
|
+
Requires-Dist: twine; extra == "dev"
|
|
38
|
+
Dynamic: license-file
|
|
39
|
+
|
|
40
|
+
<img src="https://raw.githubusercontent.com/lshpaner/pitsurf/main/docs/logo.svg" width="520" alt="pitsurf">
|
|
41
|
+
|
|
42
|
+
[](https://pypi.org/project/pitsurf/)
|
|
43
|
+
[](https://github.com/lshpaner/pitsurf/blob/main/LICENSE)
|
|
44
|
+

|
|
45
|
+
|
|
46
|
+
Surface construction from scattered survey points for open-pit mine surveying.
|
|
47
|
+
|
|
48
|
+
`pitsurf` is a Python implementation of the surface-modeling methods used in
|
|
49
|
+
the automated geological and mine-surveying support system developed at the
|
|
50
|
+
State University of Economics and Technology (Kryvyi Rih, Ukraine), with a
|
|
51
|
+
kernel-based interpolator added for comparison. All methods share one
|
|
52
|
+
`fit` / `predict` interface.
|
|
53
|
+
|
|
54
|
+
> **Status:** pre-alpha. The API will change. Nothing here is ready for
|
|
55
|
+
> production use yet.
|
|
56
|
+
|
|
57
|
+
## Methods
|
|
58
|
+
|
|
59
|
+
| Class | Method | Source |
|
|
60
|
+
|---|---|---|
|
|
61
|
+
| `NearestRegion` | Polygonal (nearest-region / Thiessen) surface | Zelensky & Lysenko (2023) |
|
|
62
|
+
| `SectorIDW` | Sector-constrained inverse distance weighting, optional anisotropy via critical correlation radius | Zelenskyi et al. (2020); Zelensky & Donchenko (2025) |
|
|
63
|
+
| `PolynomialSurface` | Bivariate least-squares polynomial up to degree 5, with adequacy statistics | Zelenskyi et al. (2020); Zelensky & Donchenko (2025) |
|
|
64
|
+
| `NURBSSurface` | Tensor-product NURBS smoother over an interpolated grid | Zelensky & Lysenko (2023) |
|
|
65
|
+
| `KernelSurface` | SVR / Gaussian process with the critical correlation radius as default length scale | this package |
|
|
66
|
+
|
|
67
|
+
Utilities: `acf_radius()` (critical correlation radius from the spatial
|
|
68
|
+
autocorrelation function) and `prism_volume()` (volume between surfaces by
|
|
69
|
+
unit prisms).
|
|
70
|
+
|
|
71
|
+
Implemented so far: `NearestRegion`, `SectorIDW`.
|
|
72
|
+
|
|
73
|
+
## Install
|
|
74
|
+
|
|
75
|
+
```bash
|
|
76
|
+
pip install pitsurf
|
|
77
|
+
```
|
|
78
|
+
|
|
79
|
+
## Usage
|
|
80
|
+
|
|
81
|
+
```python
|
|
82
|
+
import numpy as np
|
|
83
|
+
from pitsurf import NearestRegion
|
|
84
|
+
|
|
85
|
+
xy = np.array([[0.0, 0.0], [10.0, 0.0], [0.0, 10.0], [10.0, 10.0]])
|
|
86
|
+
z = np.array([100.0, 102.0, 101.0, 103.0])
|
|
87
|
+
|
|
88
|
+
model = NearestRegion().fit(xy, z)
|
|
89
|
+
model.predict(np.array([[2.0, 1.0], [9.0, 9.0]]))
|
|
90
|
+
# array([100., 103.])
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
## Repository layout
|
|
94
|
+
|
|
95
|
+
```
|
|
96
|
+
pitsurf/ package; core/ holds the interpolators
|
|
97
|
+
data_gen.py fetch a raw DEM tile -> data/raw (Parquet, optional CSV)
|
|
98
|
+
preprocessing.py subsample to sparse survey points -> data/processed
|
|
99
|
+
notebooks/ analysis; outputs stripped before commit
|
|
100
|
+
data/ gitignored
|
|
101
|
+
```
|
|
102
|
+
|
|
103
|
+
## Development
|
|
104
|
+
|
|
105
|
+
```bash
|
|
106
|
+
conda activate conda_312
|
|
107
|
+
make create_venv && source venv_pit/bin/activate
|
|
108
|
+
make requirements
|
|
109
|
+
make test
|
|
110
|
+
make help # list all targets
|
|
111
|
+
```
|
|
112
|
+
|
|
113
|
+
## References
|
|
114
|
+
|
|
115
|
+
Zelenskyi, O., Lysenko, V., Alekseyev, M., & Vlasov, V. (2020). Modeling of
|
|
116
|
+
the earth's surface subsidence during its undermining by stoping in coal mines.
|
|
117
|
+
*Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu*, (6), 121–127.
|
|
118
|
+
https://doi.org/10.33271/nvngu/2020-6/121
|
|
119
|
+
|
|
120
|
+
Zelensky, O., & Lysenko, V. (2023). Construction surface using desktop and
|
|
121
|
+
mobile applications. *Economics and Technical Engineering*, 1(1), 87–102.
|
|
122
|
+
https://doi.org/10.62911/ete.2023.01.01.07
|
|
123
|
+
|
|
124
|
+
Zelensky, O., & Donchenko, O. (2025). Research and application of polynomial
|
|
125
|
+
models in the development of an automated geological and mine surveying
|
|
126
|
+
support system for open-pit mining. *Mining Journal of Kryvyi Rih National
|
|
127
|
+
University*, 59(1), 10–21. https://doi.org/10.31721/2306-5435-2025-1-10-21
|
|
128
|
+
|
|
129
|
+
## License
|
|
130
|
+
|
|
131
|
+
MIT
|
|
@@ -0,0 +1,92 @@
|
|
|
1
|
+
<img src="https://raw.githubusercontent.com/lshpaner/pitsurf/main/docs/logo.svg" width="520" alt="pitsurf">
|
|
2
|
+
|
|
3
|
+
[](https://pypi.org/project/pitsurf/)
|
|
4
|
+
[](https://github.com/lshpaner/pitsurf/blob/main/LICENSE)
|
|
5
|
+

|
|
6
|
+
|
|
7
|
+
Surface construction from scattered survey points for open-pit mine surveying.
|
|
8
|
+
|
|
9
|
+
`pitsurf` is a Python implementation of the surface-modeling methods used in
|
|
10
|
+
the automated geological and mine-surveying support system developed at the
|
|
11
|
+
State University of Economics and Technology (Kryvyi Rih, Ukraine), with a
|
|
12
|
+
kernel-based interpolator added for comparison. All methods share one
|
|
13
|
+
`fit` / `predict` interface.
|
|
14
|
+
|
|
15
|
+
> **Status:** pre-alpha. The API will change. Nothing here is ready for
|
|
16
|
+
> production use yet.
|
|
17
|
+
|
|
18
|
+
## Methods
|
|
19
|
+
|
|
20
|
+
| Class | Method | Source |
|
|
21
|
+
|---|---|---|
|
|
22
|
+
| `NearestRegion` | Polygonal (nearest-region / Thiessen) surface | Zelensky & Lysenko (2023) |
|
|
23
|
+
| `SectorIDW` | Sector-constrained inverse distance weighting, optional anisotropy via critical correlation radius | Zelenskyi et al. (2020); Zelensky & Donchenko (2025) |
|
|
24
|
+
| `PolynomialSurface` | Bivariate least-squares polynomial up to degree 5, with adequacy statistics | Zelenskyi et al. (2020); Zelensky & Donchenko (2025) |
|
|
25
|
+
| `NURBSSurface` | Tensor-product NURBS smoother over an interpolated grid | Zelensky & Lysenko (2023) |
|
|
26
|
+
| `KernelSurface` | SVR / Gaussian process with the critical correlation radius as default length scale | this package |
|
|
27
|
+
|
|
28
|
+
Utilities: `acf_radius()` (critical correlation radius from the spatial
|
|
29
|
+
autocorrelation function) and `prism_volume()` (volume between surfaces by
|
|
30
|
+
unit prisms).
|
|
31
|
+
|
|
32
|
+
Implemented so far: `NearestRegion`, `SectorIDW`.
|
|
33
|
+
|
|
34
|
+
## Install
|
|
35
|
+
|
|
36
|
+
```bash
|
|
37
|
+
pip install pitsurf
|
|
38
|
+
```
|
|
39
|
+
|
|
40
|
+
## Usage
|
|
41
|
+
|
|
42
|
+
```python
|
|
43
|
+
import numpy as np
|
|
44
|
+
from pitsurf import NearestRegion
|
|
45
|
+
|
|
46
|
+
xy = np.array([[0.0, 0.0], [10.0, 0.0], [0.0, 10.0], [10.0, 10.0]])
|
|
47
|
+
z = np.array([100.0, 102.0, 101.0, 103.0])
|
|
48
|
+
|
|
49
|
+
model = NearestRegion().fit(xy, z)
|
|
50
|
+
model.predict(np.array([[2.0, 1.0], [9.0, 9.0]]))
|
|
51
|
+
# array([100., 103.])
|
|
52
|
+
```
|
|
53
|
+
|
|
54
|
+
## Repository layout
|
|
55
|
+
|
|
56
|
+
```
|
|
57
|
+
pitsurf/ package; core/ holds the interpolators
|
|
58
|
+
data_gen.py fetch a raw DEM tile -> data/raw (Parquet, optional CSV)
|
|
59
|
+
preprocessing.py subsample to sparse survey points -> data/processed
|
|
60
|
+
notebooks/ analysis; outputs stripped before commit
|
|
61
|
+
data/ gitignored
|
|
62
|
+
```
|
|
63
|
+
|
|
64
|
+
## Development
|
|
65
|
+
|
|
66
|
+
```bash
|
|
67
|
+
conda activate conda_312
|
|
68
|
+
make create_venv && source venv_pit/bin/activate
|
|
69
|
+
make requirements
|
|
70
|
+
make test
|
|
71
|
+
make help # list all targets
|
|
72
|
+
```
|
|
73
|
+
|
|
74
|
+
## References
|
|
75
|
+
|
|
76
|
+
Zelenskyi, O., Lysenko, V., Alekseyev, M., & Vlasov, V. (2020). Modeling of
|
|
77
|
+
the earth's surface subsidence during its undermining by stoping in coal mines.
|
|
78
|
+
*Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu*, (6), 121–127.
|
|
79
|
+
https://doi.org/10.33271/nvngu/2020-6/121
|
|
80
|
+
|
|
81
|
+
Zelensky, O., & Lysenko, V. (2023). Construction surface using desktop and
|
|
82
|
+
mobile applications. *Economics and Technical Engineering*, 1(1), 87–102.
|
|
83
|
+
https://doi.org/10.62911/ete.2023.01.01.07
|
|
84
|
+
|
|
85
|
+
Zelensky, O., & Donchenko, O. (2025). Research and application of polynomial
|
|
86
|
+
models in the development of an automated geological and mine surveying
|
|
87
|
+
support system for open-pit mining. *Mining Journal of Kryvyi Rih National
|
|
88
|
+
University*, 59(1), 10–21. https://doi.org/10.31721/2306-5435-2025-1-10-21
|
|
89
|
+
|
|
90
|
+
## License
|
|
91
|
+
|
|
92
|
+
MIT
|
|
File without changes
|
|
@@ -0,0 +1,160 @@
|
|
|
1
|
+
"""Interpolators that share one fit/predict interface.
|
|
2
|
+
|
|
3
|
+
Each class takes survey points as an (n, 2) array of x, y coordinates and an
|
|
4
|
+
(n,) array of elevations (or any indicator such as ore grade) z, and predicts
|
|
5
|
+
z at new x, y locations.
|
|
6
|
+
|
|
7
|
+
Methods follow the automated geological and mine-surveying support system
|
|
8
|
+
described in Zelenskyi et al. (2020), Zelensky & Lysenko (2023), and
|
|
9
|
+
Zelensky & Donchenko (2025).
|
|
10
|
+
"""
|
|
11
|
+
|
|
12
|
+
from __future__ import annotations
|
|
13
|
+
|
|
14
|
+
import numpy as np
|
|
15
|
+
from scipy.interpolate import NearestNDInterpolator
|
|
16
|
+
from scipy.spatial import cKDTree
|
|
17
|
+
|
|
18
|
+
|
|
19
|
+
def _check_xy_z(xy: np.ndarray, z: np.ndarray) -> tuple[np.ndarray, np.ndarray]:
|
|
20
|
+
xy = np.asarray(xy, dtype=float)
|
|
21
|
+
z = np.asarray(z, dtype=float)
|
|
22
|
+
if xy.ndim != 2 or xy.shape[1] != 2:
|
|
23
|
+
raise ValueError("xy must have shape (n, 2)")
|
|
24
|
+
if z.shape != (xy.shape[0],):
|
|
25
|
+
raise ValueError("z must have shape (n,)")
|
|
26
|
+
return xy, z
|
|
27
|
+
|
|
28
|
+
|
|
29
|
+
def _check_query(xy: np.ndarray) -> np.ndarray:
|
|
30
|
+
xy = np.asarray(xy, dtype=float)
|
|
31
|
+
if xy.ndim != 2 or xy.shape[1] != 2:
|
|
32
|
+
raise ValueError("query xy must have shape (m, 2)")
|
|
33
|
+
return xy
|
|
34
|
+
|
|
35
|
+
|
|
36
|
+
class NearestRegion:
|
|
37
|
+
"""Polygonal (nearest-region / Thiessen polygon) surface.
|
|
38
|
+
|
|
39
|
+
Every location takes the elevation of its closest survey point, so the
|
|
40
|
+
surface is piecewise constant over the Voronoi cell of each point. This is
|
|
41
|
+
the classical polygonal method of mine surveying and reserve estimation
|
|
42
|
+
(Zelensky & Lysenko, 2023, section 4).
|
|
43
|
+
"""
|
|
44
|
+
|
|
45
|
+
def __init__(self) -> None:
|
|
46
|
+
self._interp: NearestNDInterpolator | None = None
|
|
47
|
+
|
|
48
|
+
def fit(self, xy: np.ndarray, z: np.ndarray) -> "NearestRegion":
|
|
49
|
+
xy, z = _check_xy_z(xy, z)
|
|
50
|
+
self._interp = NearestNDInterpolator(xy, z)
|
|
51
|
+
return self
|
|
52
|
+
|
|
53
|
+
def predict(self, xy: np.ndarray) -> np.ndarray:
|
|
54
|
+
if self._interp is None:
|
|
55
|
+
raise RuntimeError("call fit() before predict()")
|
|
56
|
+
xy = _check_query(xy)
|
|
57
|
+
return self._interp(xy[:, 0], xy[:, 1])
|
|
58
|
+
|
|
59
|
+
|
|
60
|
+
class SectorIDW:
|
|
61
|
+
"""Sector-constrained inverse distance weighting.
|
|
62
|
+
|
|
63
|
+
The value at a query point is a weighted average of nearby survey points,
|
|
64
|
+
with weight 1 / d**power (Zelenskyi et al., 2020; Zelensky & Donchenko,
|
|
65
|
+
2025, eq. 11). To stop a cluster of points on one side from dominating,
|
|
66
|
+
the search circle of the given radius is cut into ``n_sectors`` equal
|
|
67
|
+
angular sectors and only the single closest point in each sector is used
|
|
68
|
+
(2020 paper, Fig. 3). Query points with no survey point inside the radius
|
|
69
|
+
return NaN.
|
|
70
|
+
|
|
71
|
+
Parameters
|
|
72
|
+
----------
|
|
73
|
+
radius : float
|
|
74
|
+
Search radius in the units of xy.
|
|
75
|
+
power : float, default 2.0
|
|
76
|
+
Distance exponent alpha. 2 is the usual choice.
|
|
77
|
+
n_sectors : int, default 8
|
|
78
|
+
Number of angular sectors around each query point.
|
|
79
|
+
anisotropy : tuple[float, float] or None
|
|
80
|
+
Critical correlation radius along x and along y, (rk_x, rk_y). When
|
|
81
|
+
given, distances are scaled by rk_max / rk_axis before weighting so a
|
|
82
|
+
point along the better-correlated axis counts as closer (Zelensky &
|
|
83
|
+
Donchenko, 2025, eq. 12-13). None means isotropic.
|
|
84
|
+
"""
|
|
85
|
+
|
|
86
|
+
def __init__(
|
|
87
|
+
self,
|
|
88
|
+
radius: float,
|
|
89
|
+
power: float = 2.0,
|
|
90
|
+
n_sectors: int = 8,
|
|
91
|
+
anisotropy: tuple[float, float] | None = None,
|
|
92
|
+
) -> None:
|
|
93
|
+
if radius <= 0:
|
|
94
|
+
raise ValueError("radius must be positive")
|
|
95
|
+
if n_sectors < 1:
|
|
96
|
+
raise ValueError("n_sectors must be at least 1")
|
|
97
|
+
if anisotropy is not None:
|
|
98
|
+
rk_x, rk_y = anisotropy
|
|
99
|
+
if rk_x <= 0 or rk_y <= 0:
|
|
100
|
+
raise ValueError("anisotropy radii must be positive")
|
|
101
|
+
self.radius = float(radius)
|
|
102
|
+
self.power = float(power)
|
|
103
|
+
self.n_sectors = int(n_sectors)
|
|
104
|
+
self.anisotropy = anisotropy
|
|
105
|
+
self._xy: np.ndarray | None = None
|
|
106
|
+
self._z: np.ndarray | None = None
|
|
107
|
+
self._tree: cKDTree | None = None
|
|
108
|
+
|
|
109
|
+
# scale factors that stretch the axis with the shorter correlation radius
|
|
110
|
+
def _axis_scale(self) -> np.ndarray:
|
|
111
|
+
if self.anisotropy is None:
|
|
112
|
+
return np.ones(2)
|
|
113
|
+
rk = np.asarray(self.anisotropy, dtype=float)
|
|
114
|
+
return rk.max() / rk
|
|
115
|
+
|
|
116
|
+
def fit(self, xy: np.ndarray, z: np.ndarray) -> "SectorIDW":
|
|
117
|
+
xy, z = _check_xy_z(xy, z)
|
|
118
|
+
self._xy = xy
|
|
119
|
+
self._z = z
|
|
120
|
+
# the tree is built on scaled coordinates so the radius search already
|
|
121
|
+
# respects anisotropy
|
|
122
|
+
self._tree = cKDTree(xy * self._axis_scale())
|
|
123
|
+
return self
|
|
124
|
+
|
|
125
|
+
def _predict_one(self, q: np.ndarray) -> float:
|
|
126
|
+
assert self._tree is not None and self._xy is not None and self._z is not None
|
|
127
|
+
scale = self._axis_scale()
|
|
128
|
+
q_scaled = q * scale
|
|
129
|
+
idx = self._tree.query_ball_point(q_scaled, self.radius)
|
|
130
|
+
if len(idx) == 0:
|
|
131
|
+
return np.nan
|
|
132
|
+
idx = np.asarray(idx)
|
|
133
|
+
|
|
134
|
+
delta = self._xy[idx] * scale - q_scaled # (k, 2) offsets, scaled
|
|
135
|
+
d = np.hypot(delta[:, 0], delta[:, 1])
|
|
136
|
+
|
|
137
|
+
# a survey point sitting exactly on the query location wins outright
|
|
138
|
+
hit = np.where(d == 0.0)[0]
|
|
139
|
+
if hit.size:
|
|
140
|
+
return float(self._z[idx[hit[0]]])
|
|
141
|
+
|
|
142
|
+
# assign each candidate to an angular sector and keep the closest per sector
|
|
143
|
+
angle = np.arctan2(delta[:, 1], delta[:, 0]) # (-pi, pi]
|
|
144
|
+
sector = np.floor((angle + np.pi) / (2.0 * np.pi / self.n_sectors)).astype(int)
|
|
145
|
+
sector = np.clip(sector, 0, self.n_sectors - 1)
|
|
146
|
+
|
|
147
|
+
keep: list[int] = []
|
|
148
|
+
for s in np.unique(sector):
|
|
149
|
+
members = np.where(sector == s)[0]
|
|
150
|
+
keep.append(members[np.argmin(d[members])])
|
|
151
|
+
keep_arr = np.asarray(keep)
|
|
152
|
+
|
|
153
|
+
w = d[keep_arr] ** (-self.power)
|
|
154
|
+
return float(np.sum(w * self._z[idx[keep_arr]]) / np.sum(w))
|
|
155
|
+
|
|
156
|
+
def predict(self, xy: np.ndarray) -> np.ndarray:
|
|
157
|
+
if self._tree is None:
|
|
158
|
+
raise RuntimeError("call fit() before predict()")
|
|
159
|
+
xy = _check_query(xy)
|
|
160
|
+
return np.array([self._predict_one(q) for q in xy])
|
|
@@ -0,0 +1,131 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: pitsurf
|
|
3
|
+
Version: 0.0.0a1
|
|
4
|
+
Summary: Surface construction from scattered survey points: polygonal (nearest-region), B-spline, NURBS, and SVR interpolators for open-pit mine surveying.
|
|
5
|
+
Author: Leonid Shpaner
|
|
6
|
+
License: MIT
|
|
7
|
+
Project-URL: Homepage, https://github.com/lshpaner/pitsurf
|
|
8
|
+
Project-URL: Repository, https://github.com/lshpaner/pitsurf
|
|
9
|
+
Keywords: mine surveying,interpolation,open-pit,NURBS,IDW,surface modeling
|
|
10
|
+
Classifier: Development Status :: 2 - Pre-Alpha
|
|
11
|
+
Classifier: Intended Audience :: Science/Research
|
|
12
|
+
Classifier: License :: OSI Approved :: MIT License
|
|
13
|
+
Classifier: Programming Language :: Python :: 3
|
|
14
|
+
Classifier: Topic :: Scientific/Engineering
|
|
15
|
+
Requires-Python: >=3.10
|
|
16
|
+
Description-Content-Type: text/markdown
|
|
17
|
+
License-File: LICENSE
|
|
18
|
+
Requires-Dist: numpy>=1.24
|
|
19
|
+
Requires-Dist: scipy>=1.11
|
|
20
|
+
Provides-Extra: kernel
|
|
21
|
+
Requires-Dist: scikit-learn>=1.3; extra == "kernel"
|
|
22
|
+
Provides-Extra: nurbs
|
|
23
|
+
Requires-Dist: geomdl>=5.3; extra == "nurbs"
|
|
24
|
+
Provides-Extra: data
|
|
25
|
+
Requires-Dist: pandas>=2.0; extra == "data"
|
|
26
|
+
Requires-Dist: pyarrow>=14; extra == "data"
|
|
27
|
+
Requires-Dist: rasterio>=1.3; extra == "data"
|
|
28
|
+
Requires-Dist: typer>=0.9; extra == "data"
|
|
29
|
+
Provides-Extra: viz
|
|
30
|
+
Requires-Dist: matplotlib>=3.7; extra == "viz"
|
|
31
|
+
Provides-Extra: dev
|
|
32
|
+
Requires-Dist: pytest; extra == "dev"
|
|
33
|
+
Requires-Dist: black; extra == "dev"
|
|
34
|
+
Requires-Dist: nbstripout; extra == "dev"
|
|
35
|
+
Requires-Dist: jupyter; extra == "dev"
|
|
36
|
+
Requires-Dist: build; extra == "dev"
|
|
37
|
+
Requires-Dist: twine; extra == "dev"
|
|
38
|
+
Dynamic: license-file
|
|
39
|
+
|
|
40
|
+
<img src="https://raw.githubusercontent.com/lshpaner/pitsurf/main/docs/logo.svg" width="520" alt="pitsurf">
|
|
41
|
+
|
|
42
|
+
[](https://pypi.org/project/pitsurf/)
|
|
43
|
+
[](https://github.com/lshpaner/pitsurf/blob/main/LICENSE)
|
|
44
|
+

|
|
45
|
+
|
|
46
|
+
Surface construction from scattered survey points for open-pit mine surveying.
|
|
47
|
+
|
|
48
|
+
`pitsurf` is a Python implementation of the surface-modeling methods used in
|
|
49
|
+
the automated geological and mine-surveying support system developed at the
|
|
50
|
+
State University of Economics and Technology (Kryvyi Rih, Ukraine), with a
|
|
51
|
+
kernel-based interpolator added for comparison. All methods share one
|
|
52
|
+
`fit` / `predict` interface.
|
|
53
|
+
|
|
54
|
+
> **Status:** pre-alpha. The API will change. Nothing here is ready for
|
|
55
|
+
> production use yet.
|
|
56
|
+
|
|
57
|
+
## Methods
|
|
58
|
+
|
|
59
|
+
| Class | Method | Source |
|
|
60
|
+
|---|---|---|
|
|
61
|
+
| `NearestRegion` | Polygonal (nearest-region / Thiessen) surface | Zelensky & Lysenko (2023) |
|
|
62
|
+
| `SectorIDW` | Sector-constrained inverse distance weighting, optional anisotropy via critical correlation radius | Zelenskyi et al. (2020); Zelensky & Donchenko (2025) |
|
|
63
|
+
| `PolynomialSurface` | Bivariate least-squares polynomial up to degree 5, with adequacy statistics | Zelenskyi et al. (2020); Zelensky & Donchenko (2025) |
|
|
64
|
+
| `NURBSSurface` | Tensor-product NURBS smoother over an interpolated grid | Zelensky & Lysenko (2023) |
|
|
65
|
+
| `KernelSurface` | SVR / Gaussian process with the critical correlation radius as default length scale | this package |
|
|
66
|
+
|
|
67
|
+
Utilities: `acf_radius()` (critical correlation radius from the spatial
|
|
68
|
+
autocorrelation function) and `prism_volume()` (volume between surfaces by
|
|
69
|
+
unit prisms).
|
|
70
|
+
|
|
71
|
+
Implemented so far: `NearestRegion`, `SectorIDW`.
|
|
72
|
+
|
|
73
|
+
## Install
|
|
74
|
+
|
|
75
|
+
```bash
|
|
76
|
+
pip install pitsurf
|
|
77
|
+
```
|
|
78
|
+
|
|
79
|
+
## Usage
|
|
80
|
+
|
|
81
|
+
```python
|
|
82
|
+
import numpy as np
|
|
83
|
+
from pitsurf import NearestRegion
|
|
84
|
+
|
|
85
|
+
xy = np.array([[0.0, 0.0], [10.0, 0.0], [0.0, 10.0], [10.0, 10.0]])
|
|
86
|
+
z = np.array([100.0, 102.0, 101.0, 103.0])
|
|
87
|
+
|
|
88
|
+
model = NearestRegion().fit(xy, z)
|
|
89
|
+
model.predict(np.array([[2.0, 1.0], [9.0, 9.0]]))
|
|
90
|
+
# array([100., 103.])
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
## Repository layout
|
|
94
|
+
|
|
95
|
+
```
|
|
96
|
+
pitsurf/ package; core/ holds the interpolators
|
|
97
|
+
data_gen.py fetch a raw DEM tile -> data/raw (Parquet, optional CSV)
|
|
98
|
+
preprocessing.py subsample to sparse survey points -> data/processed
|
|
99
|
+
notebooks/ analysis; outputs stripped before commit
|
|
100
|
+
data/ gitignored
|
|
101
|
+
```
|
|
102
|
+
|
|
103
|
+
## Development
|
|
104
|
+
|
|
105
|
+
```bash
|
|
106
|
+
conda activate conda_312
|
|
107
|
+
make create_venv && source venv_pit/bin/activate
|
|
108
|
+
make requirements
|
|
109
|
+
make test
|
|
110
|
+
make help # list all targets
|
|
111
|
+
```
|
|
112
|
+
|
|
113
|
+
## References
|
|
114
|
+
|
|
115
|
+
Zelenskyi, O., Lysenko, V., Alekseyev, M., & Vlasov, V. (2020). Modeling of
|
|
116
|
+
the earth's surface subsidence during its undermining by stoping in coal mines.
|
|
117
|
+
*Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu*, (6), 121–127.
|
|
118
|
+
https://doi.org/10.33271/nvngu/2020-6/121
|
|
119
|
+
|
|
120
|
+
Zelensky, O., & Lysenko, V. (2023). Construction surface using desktop and
|
|
121
|
+
mobile applications. *Economics and Technical Engineering*, 1(1), 87–102.
|
|
122
|
+
https://doi.org/10.62911/ete.2023.01.01.07
|
|
123
|
+
|
|
124
|
+
Zelensky, O., & Donchenko, O. (2025). Research and application of polynomial
|
|
125
|
+
models in the development of an automated geological and mine surveying
|
|
126
|
+
support system for open-pit mining. *Mining Journal of Kryvyi Rih National
|
|
127
|
+
University*, 59(1), 10–21. https://doi.org/10.31721/2306-5435-2025-1-10-21
|
|
128
|
+
|
|
129
|
+
## License
|
|
130
|
+
|
|
131
|
+
MIT
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
LICENSE
|
|
2
|
+
README.md
|
|
3
|
+
pyproject.toml
|
|
4
|
+
pitsurf/__init__.py
|
|
5
|
+
pitsurf.egg-info/PKG-INFO
|
|
6
|
+
pitsurf.egg-info/SOURCES.txt
|
|
7
|
+
pitsurf.egg-info/dependency_links.txt
|
|
8
|
+
pitsurf.egg-info/requires.txt
|
|
9
|
+
pitsurf.egg-info/top_level.txt
|
|
10
|
+
pitsurf/core/__init__.py
|
|
11
|
+
pitsurf/core/interpolation.py
|
|
12
|
+
tests/test_nearest_region.py
|
|
13
|
+
tests/test_sector_idw.py
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
|
|
@@ -0,0 +1,25 @@
|
|
|
1
|
+
numpy>=1.24
|
|
2
|
+
scipy>=1.11
|
|
3
|
+
|
|
4
|
+
[data]
|
|
5
|
+
pandas>=2.0
|
|
6
|
+
pyarrow>=14
|
|
7
|
+
rasterio>=1.3
|
|
8
|
+
typer>=0.9
|
|
9
|
+
|
|
10
|
+
[dev]
|
|
11
|
+
pytest
|
|
12
|
+
black
|
|
13
|
+
nbstripout
|
|
14
|
+
jupyter
|
|
15
|
+
build
|
|
16
|
+
twine
|
|
17
|
+
|
|
18
|
+
[kernel]
|
|
19
|
+
scikit-learn>=1.3
|
|
20
|
+
|
|
21
|
+
[nurbs]
|
|
22
|
+
geomdl>=5.3
|
|
23
|
+
|
|
24
|
+
[viz]
|
|
25
|
+
matplotlib>=3.7
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
pitsurf
|
|
@@ -0,0 +1,41 @@
|
|
|
1
|
+
[build-system]
|
|
2
|
+
requires = ["setuptools>=68", "wheel"]
|
|
3
|
+
build-backend = "setuptools.build_meta"
|
|
4
|
+
|
|
5
|
+
[project]
|
|
6
|
+
name = "pitsurf"
|
|
7
|
+
version = "0.0.0a1"
|
|
8
|
+
description = "Surface construction from scattered survey points: polygonal (nearest-region), B-spline, NURBS, and SVR interpolators for open-pit mine surveying."
|
|
9
|
+
readme = "README.md"
|
|
10
|
+
requires-python = ">=3.10"
|
|
11
|
+
license = {text = "MIT"}
|
|
12
|
+
authors = [{name = "Leonid Shpaner"}]
|
|
13
|
+
keywords = ["mine surveying", "interpolation", "open-pit", "NURBS", "IDW", "surface modeling"]
|
|
14
|
+
classifiers = [
|
|
15
|
+
"Development Status :: 2 - Pre-Alpha",
|
|
16
|
+
"Intended Audience :: Science/Research",
|
|
17
|
+
"License :: OSI Approved :: MIT License",
|
|
18
|
+
"Programming Language :: Python :: 3",
|
|
19
|
+
"Topic :: Scientific/Engineering",
|
|
20
|
+
]
|
|
21
|
+
dependencies = [
|
|
22
|
+
"numpy>=1.24",
|
|
23
|
+
"scipy>=1.11",
|
|
24
|
+
]
|
|
25
|
+
|
|
26
|
+
[project.urls]
|
|
27
|
+
Homepage = "https://github.com/lshpaner/pitsurf"
|
|
28
|
+
Repository = "https://github.com/lshpaner/pitsurf"
|
|
29
|
+
|
|
30
|
+
[project.optional-dependencies]
|
|
31
|
+
kernel = ["scikit-learn>=1.3"]
|
|
32
|
+
nurbs = ["geomdl>=5.3"]
|
|
33
|
+
data = ["pandas>=2.0", "pyarrow>=14", "rasterio>=1.3", "typer>=0.9"]
|
|
34
|
+
viz = ["matplotlib>=3.7"]
|
|
35
|
+
dev = ["pytest", "black", "nbstripout", "jupyter", "build", "twine"]
|
|
36
|
+
|
|
37
|
+
[tool.setuptools.packages.find]
|
|
38
|
+
include = ["pitsurf*"]
|
|
39
|
+
|
|
40
|
+
[tool.black]
|
|
41
|
+
line-length = 88
|
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
import numpy as np
|
|
2
|
+
|
|
3
|
+
from pitsurf import NearestRegion
|
|
4
|
+
|
|
5
|
+
|
|
6
|
+
def test_nearest_region_reproduces_training_points():
|
|
7
|
+
rng = np.random.default_rng(0)
|
|
8
|
+
xy = rng.uniform(0, 100, size=(50, 2))
|
|
9
|
+
z = rng.normal(500, 10, size=50)
|
|
10
|
+
model = NearestRegion().fit(xy, z)
|
|
11
|
+
np.testing.assert_allclose(model.predict(xy), z)
|
|
12
|
+
|
|
13
|
+
|
|
14
|
+
def test_nearest_region_is_piecewise_constant():
|
|
15
|
+
xy = np.array([[0.0, 0.0], [10.0, 0.0]])
|
|
16
|
+
z = np.array([1.0, 2.0])
|
|
17
|
+
model = NearestRegion().fit(xy, z)
|
|
18
|
+
assert model.predict(np.array([[2.0, 0.0]]))[0] == 1.0
|
|
19
|
+
assert model.predict(np.array([[8.0, 0.0]]))[0] == 2.0
|
|
@@ -0,0 +1,55 @@
|
|
|
1
|
+
import numpy as np
|
|
2
|
+
import pytest
|
|
3
|
+
|
|
4
|
+
from pitsurf import SectorIDW
|
|
5
|
+
|
|
6
|
+
|
|
7
|
+
def test_exact_at_training_points():
|
|
8
|
+
rng = np.random.default_rng(1)
|
|
9
|
+
xy = rng.uniform(0, 100, size=(40, 2))
|
|
10
|
+
z = rng.normal(500, 10, size=40)
|
|
11
|
+
model = SectorIDW(radius=50).fit(xy, z)
|
|
12
|
+
np.testing.assert_allclose(model.predict(xy), z)
|
|
13
|
+
|
|
14
|
+
|
|
15
|
+
def test_symmetric_square_returns_mean_at_center():
|
|
16
|
+
# four points at equal distance from the origin, one per quadrant
|
|
17
|
+
xy = np.array([[1.0, 1.0], [-1.0, 1.0], [-1.0, -1.0], [1.0, -1.0]])
|
|
18
|
+
z = np.array([10.0, 20.0, 30.0, 40.0])
|
|
19
|
+
model = SectorIDW(radius=5, n_sectors=4).fit(xy, z)
|
|
20
|
+
assert model.predict(np.array([[0.0, 0.0]]))[0] == pytest.approx(25.0)
|
|
21
|
+
|
|
22
|
+
|
|
23
|
+
def test_outside_radius_is_nan():
|
|
24
|
+
xy = np.array([[0.0, 0.0], [1.0, 0.0]])
|
|
25
|
+
z = np.array([1.0, 2.0])
|
|
26
|
+
model = SectorIDW(radius=2).fit(xy, z)
|
|
27
|
+
assert np.isnan(model.predict(np.array([[100.0, 100.0]]))[0])
|
|
28
|
+
|
|
29
|
+
|
|
30
|
+
def test_sector_rule_limits_cluster_influence():
|
|
31
|
+
# nine points clustered just north of the origin, one point south
|
|
32
|
+
north = np.column_stack([np.linspace(-0.5, 0.5, 9), np.full(9, 2.0)])
|
|
33
|
+
south = np.array([[0.0, -2.0]])
|
|
34
|
+
xy = np.vstack([north, south])
|
|
35
|
+
z = np.concatenate([np.full(9, 100.0), [0.0]])
|
|
36
|
+
|
|
37
|
+
plain = SectorIDW(radius=10, n_sectors=1).fit(xy, z) # effectively plain IDW
|
|
38
|
+
sectored = SectorIDW(radius=10, n_sectors=8).fit(xy, z)
|
|
39
|
+
q = np.array([[0.0, 0.0]])
|
|
40
|
+
|
|
41
|
+
# plain IDW is pulled hard toward the cluster; sectoring pulls it back toward the middle
|
|
42
|
+
assert plain.predict(q)[0] > 85.0
|
|
43
|
+
assert sectored.predict(q)[0] < plain.predict(q)[0]
|
|
44
|
+
|
|
45
|
+
|
|
46
|
+
def test_anisotropy_favours_long_axis():
|
|
47
|
+
# one point east, one point north, same distance; stretch correlation along x
|
|
48
|
+
xy = np.array([[10.0, 0.0], [0.0, 10.0]])
|
|
49
|
+
z = np.array([0.0, 100.0])
|
|
50
|
+
iso = SectorIDW(radius=50).fit(xy, z)
|
|
51
|
+
aniso = SectorIDW(radius=50, anisotropy=(40.0, 10.0)).fit(xy, z)
|
|
52
|
+
q = np.array([[0.0, 0.0]])
|
|
53
|
+
assert iso.predict(q)[0] == pytest.approx(50.0)
|
|
54
|
+
# the east point lies along the long axis so it should pull the estimate down
|
|
55
|
+
assert aniso.predict(q)[0] < 50.0
|