mss-imageproc 0.0.1__tar.gz

mss_imageproc-0.0.1/PKG-INFO

@@ -0,0 +1,55 @@
+Metadata-Version: 2.4
+Name: mss-imageproc
+Version: 0.0.1
+Summary: Image processing utilities for MSS Designer Models
+Author-email: "Sunip K. Mukherjee" <sunipkmukherjee@gmail.com>
+License: MIT
+Keywords: image-processing,remote-sensing,mosaic
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+Requires-Dist: astropy>=7.1
+Requires-Dist: dacite>=1.8.1
+Requires-Dist: natsort>=8.4.0
+Requires-Dist: numpy>=1.26
+Requires-Dist: scikit-image>=0.22
+Requires-Dist: serde-dataclass>=0.0.4
+Requires-Dist: tomlkit>=0.12.5
+Requires-Dist: xarray>=2024.1.0
+Requires-Dist: netcdf4>=1.6.3
+Requires-Dist: astropy-xarray==0.1.0
+
+# mss-imageproc
+This repository contains the Python library `mss-imageproc`, which provides tools for processing images 
+from instruments modeled using the Multi-Slit Spectrograph (MSS) Designer. The library includes functions 
+applying image transformations to map detector images onto the focal plane of the instrument, as well as
+remove diffraction slit curvature from the images. This library is designed to be used stand-alone, using
+the instrument configuration files generated by the MSS Designer. This library is also used internally by
+the MSS Designer.
+
+# Installation
+You can install the library using pip:
+```bash
+pip install mss-imageproc
+```
+
+# Usage
+The library is used in the following way:
+```python
+# Import the library
+from mss_imageproc import MosaicImageStraightener
+
+# Create a straightener object using the instrument configuration file
+straightener = MosaicImageStraightener.from_instrument_config('path/to/instrument_curve_maps.bin')
+# Load an image and map it onto the mosaic grid
+image_array = ...  # Load your image as a 2D NumPy array
+mapped_image = straightener.load_image(image_array)
+# Straighten the image by removing slit curvature
+straightened_images = straightener.straighten_image(mapped_image)
+# The result is a dictionary of straightened images, one for each window. You can access them like this:
+for window_name, straightened_image in straightened_images.items():
+    print(f"Straightened image for window {window_name}:")
+    straightened_image.show()  # Show the straightened image
+```

mss_imageproc-0.0.1/README.md

@@ -0,0 +1,32 @@
+# mss-imageproc
+This repository contains the Python library `mss-imageproc`, which provides tools for processing images 
+from instruments modeled using the Multi-Slit Spectrograph (MSS) Designer. The library includes functions 
+applying image transformations to map detector images onto the focal plane of the instrument, as well as
+remove diffraction slit curvature from the images. This library is designed to be used stand-alone, using
+the instrument configuration files generated by the MSS Designer. This library is also used internally by
+the MSS Designer.
+
+# Installation
+You can install the library using pip:
+```bash
+pip install mss-imageproc
+```
+
+# Usage
+The library is used in the following way:
+```python
+# Import the library
+from mss_imageproc import MosaicImageStraightener
+
+# Create a straightener object using the instrument configuration file
+straightener = MosaicImageStraightener.from_instrument_config('path/to/instrument_curve_maps.bin')
+# Load an image and map it onto the mosaic grid
+image_array = ...  # Load your image as a 2D NumPy array
+mapped_image = straightener.load_image(image_array)
+# Straighten the image by removing slit curvature
+straightened_images = straightener.straighten_image(mapped_image)
+# The result is a dictionary of straightened images, one for each window. You can access them like this:
+for window_name, straightened_image in straightened_images.items():
+    print(f"Straightened image for window {window_name}:")
+    straightened_image.show()  # Show the straightened image
+```

mss_imageproc-0.0.1/pyproject.toml

@@ -0,0 +1,35 @@
+[build-system]
+requires = ["setuptools>=68", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "mss-imageproc"
+version = "0.0.1"
+description = "Image processing utilities for MSS Designer Models"
+readme = "README.md"
+requires-python = ">=3.12"
+authors = [{ name = "Sunip K. Mukherjee", email = "sunipkmukherjee@gmail.com" }]
+keywords = ["image-processing", "remote-sensing", "mosaic"]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.12",
+    "Operating System :: OS Independent",
+]
+dependencies = [
+    "astropy>=7.1",
+    "dacite>=1.8.1",
+    "natsort>=8.4.0",
+    "numpy>=1.26",
+    "scikit-image>=0.22",
+    "serde-dataclass>=0.0.4",
+    "tomlkit>=0.12.5",
+    "xarray>=2024.1.0",
+    "netcdf4>=1.6.3",
+    "astropy-xarray==0.1.0",
+]
+license = { text = "MIT" }
+[tool.setuptools]
+package-dir = { "" = "src" }
+
+[tool.setuptools.packages.find]
+where = ["src"]

mss_imageproc-0.0.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

mss_imageproc-0.0.1/src/mss_imageproc/__init__.py

@@ -0,0 +1,22 @@
+from importlib.metadata import version
+
+__version__ = version(__package__ or 'mss_imageproc')
+
+from .straighten_image import (
+    MosaicImageMapper,
+    MosaicImageStraightener,
+    ScaleType,
+    TranslationType,
+    PixelSizeType,
+    TransformationMatrix
+)
+
+__all__ = [
+    "MosaicImageMapper",
+    "MosaicImageStraightener",
+    "ScaleType",
+    "TranslationType",
+    "PixelSizeType",
+    "TransformationMatrix",
+    "__version__",
+]

mss_imageproc-0.0.1/src/mss_imageproc/straighten_image.py

@@ -0,0 +1,522 @@
+# %% Imports
+from __future__ import annotations
+from json import JSONEncoder
+from pathlib import Path
+import tarfile
+from tempfile import TemporaryDirectory
+from typing import Any, Dict, List, Literal, Optional, Sequence, Tuple, Union, overload
+from dacite import Config
+from tomlkit import register_encoder
+from tomlkit.items import Item as TomlItem, item as tomlitem
+from xarray import Dataset, DataArray, concat, MergeError, load_dataset
+from astropy.units import Quantity
+import astropy.units as u
+from skimage.transform import warp, AffineTransform
+from natsort import natsorted
+from dataclasses import dataclass, field
+from numpy import arange, asarray, fromstring, interp, meshgrid, sqrt, stack, nan, ndarray, where
+from numpy.typing import NDArray
+from serde_dataclass import json_config, toml_config, JsonDataclass, TomlDataclass
+import astropy_xarray as _
+import sys
+
+# %% Type Aliases
+if sys.version_info >= (3, 11):
+    type MaybeQuantity = str | Quantity
+else:
+    from typing import TypeAlias, Union
+    MaybeQuantity: TypeAlias = Union[str, Quantity]
+
+
+def to_quantity(value: MaybeQuantity) -> Quantity:
+    if isinstance(value, str):
+        return Quantity(value)
+    elif isinstance(value, Quantity):
+        return value
+    else:
+        raise ValueError('Invalid quantity specification.')
+
+
+def optional_quantity(value: Optional[MaybeQuantity]) -> Optional[Quantity]:
+    if value is None:
+        return None
+    return to_quantity(value)
+
+# %% Serde Helpers
+
+
+@register_encoder
+def qty_ndarray_encoder(obj: Any, /, _parent=None, _sort_keys=False) -> TomlItem:
+    if isinstance(obj, Quantity):
+        return tomlitem(f'{obj}')
+    elif isinstance(obj, ndarray):
+        return tomlitem(obj.tolist())
+    else:
+        raise TypeError(
+            f'Object of type {type(obj)} is not JSON serializable.')
+
+
+class QuantityEncoder(JSONEncoder):
+    def default(self, o: Any) -> Any:
+        if isinstance(o, Quantity):
+            return f'{o}'
+        elif isinstance(o, ndarray):
+            return o.tolist()
+        else:
+            return super().default(o)
+
+
+class QuantityDecoder:
+    @staticmethod
+    def decode_qty(value: str) -> Quantity:
+        value = value.strip()
+        if value.startswith('['):
+            arr_str, unit = value.rsplit(']', 1)
+            arr_str = arr_str.strip('[]')
+            arr = fromstring(arr_str, sep=' ', dtype=float)
+            return Quantity(arr, unit.strip())
+        return Quantity(value)
+
+    @staticmethod
+    def decode_ndarray(value: List[Any]) -> NDArray:
+        return asarray(value, dtype=float)
+
+    @property
+    def config(self) -> Config:
+        return Config(
+            type_hooks={
+                Quantity: self.decode_qty,
+                ndarray: self.decode_ndarray
+            },
+        )
+
+
+QUANTITY_DECODER = QuantityDecoder().config
+
+# %% Definitions
+
+ScaleType = Union[float, Tuple[float, float]]
+TranslationType = Tuple[float, float]
+PixelSizeType = Tuple[float, float]
+PaddingMode = Literal['constant', 'edge', 'symmetric', 'reflect', 'wrap']
+
+
+@dataclass
+@json_config(ser=QuantityEncoder, de=QUANTITY_DECODER)
+@toml_config(de=QUANTITY_DECODER)
+class TransformationMatrix(JsonDataclass, TomlDataclass):
+    """Reusable affine transform state and composition helper."""
+
+    matrix: ndarray = field(
+        default_factory=lambda: asarray([
+            [1.0, 0.0, 0.0],
+            [0.0, 1.0, 0.0],
+            [0.0, 0.0, 1.0],
+        ], dtype=float),
+        metadata={
+            'description': '3x3 affine transformation matrix in homogeneous coordinates.',
+            'typecheck': lambda x, _: isinstance(x, (list, ndarray)) and asarray(x).shape == (3, 3),
+        }
+    )
+
+    def __post_init__(self) -> None:
+        self.matrix = asarray(self.matrix, dtype=float)
+        if self.matrix.shape != (3, 3):
+            raise ValueError('matrix must have shape (3, 3)')
+
+    @classmethod
+    def from_matrix(
+            cls,
+            matrix: NDArray,
+    ) -> TransformationMatrix:
+        return cls(matrix=asarray(matrix, dtype=float))
+
+    def append(self, affine: AffineTransform) -> None:
+        self.matrix = affine.params @ self.matrix
+
+    def reset(self) -> None:
+        self.matrix = asarray([
+            [1.0, 0.0, 0.0],
+            [0.0, 1.0, 0.0],
+            [0.0, 0.0, 1.0],
+        ], dtype=float)
+
+    def affine(self) -> AffineTransform:
+        return AffineTransform(matrix=self.matrix.copy())
+
+    def effective_scale(self) -> Tuple[float, float]:
+        a = float(self.matrix[0, 0])
+        b = float(self.matrix[0, 1])
+        d = float(self.matrix[1, 0])
+        e = float(self.matrix[1, 1])
+        sx = float(sqrt(a * a + d * d))
+        sy = float(sqrt(b * b + e * e))
+        return abs(sx), abs(sy)
+
+
+@dataclass
+@json_config(ser=QuantityEncoder, de=QUANTITY_DECODER)
+@toml_config(de=QUANTITY_DECODER)
+class MosaicImageMapper(JsonDataclass):
+    """Map an image onto mosaic coordinates using a provided affine matrix.
+
+    This helper is intentionally lightweight compared with ``MosaicImageTransform``:
+    it requires only source image axes, mosaic bounds, and a transformation matrix.
+    """
+
+    source_x: ndarray
+    source_y: ndarray
+    target_x: ndarray
+    target_y: ndarray
+    pixel_size: PixelSizeType
+    bounds_x: Tuple[float, float]
+    bounds_y: Tuple[float, float]
+    transform: TransformationMatrix = field(
+        default_factory=TransformationMatrix)
+    _source_x0: float = field(init=False)
+    _source_y0: float = field(init=False)
+    _inv_dx: float = field(init=False)
+    _inv_dy: float = field(init=False)
+    _use_linear_x: bool = field(init=False)
+    _use_linear_y: bool = field(init=False)
+
+    def __post_init__(self) -> None:
+        self.source_x = asarray(self.source_x, dtype=float)
+        self.source_y = asarray(self.source_y, dtype=float)
+        self.target_x = asarray(self.target_x, dtype=float)
+        self.target_y = asarray(self.target_y, dtype=float)
+        if self.source_x.ndim != 1 or self.source_y.ndim != 1:
+            raise ValueError('source_x and source_y must be 1D arrays')
+        if self.target_x.ndim != 1 or self.target_y.ndim != 1:
+            raise ValueError('target_x and target_y must be 1D arrays')
+        if self.target_x.size == 0 or self.target_y.size == 0:
+            raise ValueError('target_x and target_y must not be empty')
+        if self.pixel_size[0] <= 0 or self.pixel_size[1] <= 0:
+            raise ValueError('pixel_size must be positive')
+        if not isinstance(self.transform, TransformationMatrix):
+            self.transform = TransformationMatrix.from_matrix(
+                asarray(self.transform, dtype=float))
+
+        # Match MosaicImageTransform coordinate-index behavior for consistency.
+        self._source_x0 = float(self.source_x[0])
+        self._source_y0 = float(self.source_y[0])
+        self._inv_dx = 0.0
+        self._inv_dy = 0.0
+        self._use_linear_x = False
+        self._use_linear_y = False
+
+        if self.source_x.size >= 2:
+            dx = float(self.source_x[1] - self.source_x[0])
+            if dx != 0.0:
+                xdiff = asarray(
+                    self.source_x[1:] - self.source_x[:-1], dtype=float)
+                xtol = max(1e-12, 1e-9 * abs(dx))
+                self._use_linear_x = bool((abs(xdiff - dx) <= xtol).all())
+                if self._use_linear_x:
+                    self._inv_dx = 1.0 / dx
+
+        if self.source_y.size >= 2:
+            dy = float(self.source_y[1] - self.source_y[0])
+            if dy != 0.0:
+                ydiff = asarray(
+                    self.source_y[1:] - self.source_y[:-1], dtype=float)
+                ytol = max(1e-12, 1e-9 * abs(dy))
+                self._use_linear_y = bool((abs(ydiff - dy) <= ytol).all())
+                if self._use_linear_y:
+                    self._inv_dy = 1.0 / dy
+
+    def map_to_mosaic(
+        self,
+        image: NDArray,
+        order: int = 1,
+        cval: float = nan,
+        mode: str = 'constant',
+    ) -> Tuple[DataArray, PixelSizeType]:
+        """Render a 2D image onto the finalized full-resolution mosaic grid."""
+        image_data = asarray(image)
+        if image_data.ndim != 2:
+            raise ValueError('image must be a 2D array')
+        if self.source_x.size != image_data.shape[1]:
+            raise ValueError('source_x size must match image width')
+        if self.source_y.size != image_data.shape[0]:
+            raise ValueError('source_y size must match image height')
+
+        x_target = self.target_x
+        y_target = self.target_y
+        px, py = float(self.pixel_size[0]), float(self.pixel_size[1])
+
+        xx, yy = meshgrid(x_target, y_target)
+        inv = self.transform.affine().inverse.params
+        src_x = inv[0, 0] * xx + inv[0, 1] * yy + inv[0, 2]
+        src_y = inv[1, 0] * xx + inv[1, 1] * yy + inv[1, 2]
+
+        if self._use_linear_x:
+            col = self._coord_to_index_linear(
+                src_x, self._source_x0, self._inv_dx, self.source_x.size)
+        else:
+            col = self._coord_to_index(src_x, self.source_x)
+
+        if self._use_linear_y:
+            row = self._coord_to_index_linear(
+                src_y, self._source_y0, self._inv_dy, self.source_y.size)
+        else:
+            row = self._coord_to_index(src_y, self.source_y)
+        coords = stack((row, col), axis=0)
+
+        warped = warp(
+            image_data,
+            coords,
+            order=order,
+            cval=cval,
+            mode=mode,
+            preserve_range=True,
+        )
+        out = DataArray(
+            warped,
+            dims=('y', 'x'),
+            coords={
+                'x': ('x', Quantity(x_target, u.mm), {'units': u.mm, 'description': 'Mosaic X coordinate'}),
+                'y': ('y', Quantity(y_target, u.mm), {'units': u.mm, 'description': 'Mosaic Y coordinate'}),
+            },
+        ).astropy.quantify()
+        out.attrs['pixel_scale_x_mm_per_px'] = float(px)
+        out.attrs['pixel_scale_y_mm_per_px'] = float(py)
+        out.attrs['bounds_x_mm'] = self.bounds_x
+        out.attrs['bounds_y_mm'] = self.bounds_y
+        return out, (float(px), float(py))
+
+    @staticmethod
+    def _coord_to_index(coord: NDArray, axis_values: NDArray) -> NDArray:
+        """Map physical coordinates to floating pixel indices by linear interpolation.
+
+        The source axis may be ascending or descending.
+        Coordinates outside axis range map to ``-1`` and are handled by ``warp``
+        according to the selected boundary ``mode`` and ``cval``.
+        """
+        idx = arange(axis_values.size, dtype=float)
+        axis = asarray(axis_values, dtype=float)
+        if axis.size < 2:
+            return interp(coord, axis, idx, left=-1.0, right=-1.0)
+        if axis[0] > axis[-1]:
+            axis = axis[::-1]
+            idx = idx[::-1]
+        return interp(coord, axis, idx, left=-1.0, right=-1.0)
+
+    @staticmethod
+    def _coord_to_index_linear(
+            coord: NDArray,
+            axis0: float,
+            inv_step: float,
+            size: int,
+    ) -> NDArray:
+        """Map physical coordinates to floating indices for a uniform axis.
+
+        Coordinates outside the axis extent are mapped to ``-1`` so ``warp``
+        applies the configured boundary behavior.
+        """
+        # Fast O(1) index computation: idx = (coord - axis0) / step
+        idx = (coord - axis0) * inv_step
+        # Out-of-bounds indices are masked to -1, triggering warp's boundary mode.
+        return asarray(
+            where((idx < 0.0) | (idx > (size - 1)), -1.0, idx),
+            dtype=float,
+        )
+
+
+class MosaicImageStraightener:
+    """Straighten images mapped onto the mosaic grid using curve maps associated with window names.
+    """
+
+    def __init__(
+            self,
+            imaps: Dict[str, List[Dataset]],
+            mapper: MosaicImageMapper,
+    ):
+        self._mapper = mapper
+        self._imaps = imaps
+        self._windows = list(imaps.keys())
+
+    @classmethod
+    def load(
+        cls,
+        archive_path: Path,
+    ) -> MosaicImageStraightener:
+        """Load a MosaicImageStraightener from a binary bundle containing a MosaicImageMapper and associated curve maps.
+
+        Args:
+            archive_path (Path): The path to the archive file containing the mosaic image straightener data.
+
+        Raises:
+            ValueError: If the archive file does not exist or if required data is missing.
+            ValueError: If the mosaic image mapper is not found in the archive.
+
+        Returns:
+            MosaicImageStraightener: The loaded mosaic image straightener.
+        """
+        if not archive_path.exists():
+            raise ValueError(f'Archive {archive_path} does not exist')
+        mapper = None
+        with tarfile.open(archive_path, "r:xz") as tar:
+            with TemporaryDirectory() as tmpdir:
+                tmpdir = Path(tmpdir)
+                tmpdir.mkdir(exist_ok=True, parents=True)
+                tar.extractall(path=tmpdir)
+                imaps = {}
+                for dsdir in tmpdir.iterdir():
+                    if not dsdir.is_dir():
+                        # Load the mapper
+                        if dsdir.suffix == '.json' and '_mapper' in dsdir.stem:
+                            mapper = MosaicImageMapper.from_json(dsdir.read_text())
+                    else:
+                        # Load the curve maps
+                        win_name = dsdir.name
+                        datasets = []
+                        for dsfile in natsorted(dsdir.iterdir()):
+                            if not dsfile.is_file() or dsfile.suffix != '.nc':
+                                continue
+                            ds = load_dataset(dsfile)
+                            datasets.append(ds)
+                        imaps[win_name] = datasets
+                if mapper is None:
+                    raise ValueError('MosaicImageMapper not found in archive')
+                return cls(imaps, mapper)
+
+    @property
+    def windows(self) -> List[str]:
+        """Return the list of available window names for straightening.
+
+        Returns:
+            List[str]: The list of available window names for straightening.
+        """
+        return self._windows
+
+    def load_image(
+        self,
+        image: NDArray,
+        order: int = 1,
+        cval: float = nan,
+        mode: PaddingMode = 'constant',
+    ) -> DataArray:
+        """Load an image onto the mosaic grid using the mapper, preparing it for straightening.
+
+        Args:
+            image (NDArray): The input image to be mapped onto the mosaic grid. Must be a 2D array.
+            order (int, optional): The interpolation order to use. Defaults to 1 (bilinear). Available options are 0 (nearest), 1 (bilinear), 2 (biquadratic), 3 (bicubic), 4 (biquartic), and 5 (biquintic).
+            cval (float, optional): The constant value to use for padding. Defaults to nan.
+            mode (PaddingMode, optional): The padding mode to use. Defaults to 'constant'.
+
+        Returns:
+            DataArray: The image mapped onto the mosaic grid.
+        """
+        da, _ = self._mapper.map_to_mosaic(
+            image, order, cval, mode)
+        return da
+
+    @overload
+    def straighten_image(
+        self,
+        image: DataArray,
+        window: str,
+        *,
+        inplace: bool
+    ) -> DataArray: ...
+
+    @overload
+    def straighten_image(
+        self,
+        image: DataArray,
+        window: List[str],
+        *,
+        inplace: bool
+    ) -> Dict[str, DataArray]: ...
+
+    def straighten_image(
+        self,
+        image: DataArray,
+        window: str | List[str],
+        *,
+        inplace: bool = True
+    ) -> DataArray | Dict[str, DataArray]:
+        """Straighten the given image using the curve maps associated with the specified window name(s).
+
+        Args:
+            image (DataArray): The input image to be straightened, already mapped onto the mosaic grid.
+            window (str | Sequence[str]): The name(s) of the window(s) to use for straightening.
+            inplace (bool, optional): If True, the input image will be modified in place. Defaults to True.
+
+        Raises:
+            ValueError: If the straightener is not properly initialized or if window is invalid.
+
+        Returns:
+            DataArray | Dict[str, DataArray]: _description_
+        """
+        if self._imaps is None:
+            raise ValueError('Must setup first')
+        if isinstance(window, str):
+            ret: List = []
+            for ds in self._imaps.get(window, []):
+                # Trick to select the same location across different datasets
+                imageset = Dataset(
+                    data_vars={
+                        'image': (('y', 'x'), image.values),
+                        'loc': (('y', 'x'), ds['loc'].data, ds['loc'].attrs),
+                    },
+                    coords={
+                        'y': (('y',), image['y'].values, {'units': u.mm, 'description': 'Mosaic coordinate, increasing from the bottom.'}),
+                        'x': (('x',), image['x'].values, {'units': u.mm, 'description': 'Mosaic coordinate, increasing from the left.'}),
+                    },
+                ).astropy.quantify()
+                xran = ds.attrs['xran']
+                yran = ds.attrs['yran']
+                xran = (Quantity(xran[0], u.mm), Quantity(xran[1], u.mm))
+                yran = (Quantity(yran[0], u.mm), Quantity(yran[1], u.mm))
+                xform = ds['xform'].data
+                res = ds['resolution']
+                data = imageset.where(imageset['loc'], drop=True)['image']
+                data = data.sel(y=slice(*yran), x=slice(*xran))
+                if inplace:
+                    try:
+                        data /= res.data
+                    except (MergeError, ValueError):
+                        data = data.data / res.data
+                else:
+                    data = data / res.data
+                data = warp(
+                    data.values[:, :], xform, cval=nan)
+                out = DataArray(data*10, coords={
+                    'y': (
+                        ('y',),
+                        ds['wly'].data,  # type: ignore
+                        {
+                            'units': 'mm',
+                            'description': 'Height in the mosaic coordinate, increasing from the bottom.'
+                        }
+                    ),
+                    'wavelength': (
+                        'wavelength',
+                        ds['wavelength'].data / 10.0,  # type: ignore
+                        {
+                            'units': 'nm',
+                            'description': 'Wavelength in nanometers',
+                        }
+                    ),
+                })
+                ret.append(out)
+            out: DataArray = concat(ret, dim='y', join='outer')  # type: ignore
+            out = out.sortby('y')
+            out = out.sortby('wavelength')
+            if image.attrs.get('units') is not None:
+                out.attrs['units'] = image.attrs['units'] + ' / nm'
+            else:
+                out.attrs['units'] = '1 / nm'
+            return out
+        elif isinstance(window, Sequence):
+            if len(window) == 0:
+                window = self.windows
+            return {
+                name: self.straighten_image(image, name, inplace=inplace)
+                for name in window
+            }
+        else:
+            raise ValueError(
+                'win_name must be a string or an iterable of strings')

mss_imageproc-0.0.1/src/mss_imageproc.egg-info/PKG-INFO

@@ -0,0 +1,55 @@
+Metadata-Version: 2.4
+Name: mss-imageproc
+Version: 0.0.1
+Summary: Image processing utilities for MSS Designer Models
+Author-email: "Sunip K. Mukherjee" <sunipkmukherjee@gmail.com>
+License: MIT
+Keywords: image-processing,remote-sensing,mosaic
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+Requires-Dist: astropy>=7.1
+Requires-Dist: dacite>=1.8.1
+Requires-Dist: natsort>=8.4.0
+Requires-Dist: numpy>=1.26
+Requires-Dist: scikit-image>=0.22
+Requires-Dist: serde-dataclass>=0.0.4
+Requires-Dist: tomlkit>=0.12.5
+Requires-Dist: xarray>=2024.1.0
+Requires-Dist: netcdf4>=1.6.3
+Requires-Dist: astropy-xarray==0.1.0
+
+# mss-imageproc
+This repository contains the Python library `mss-imageproc`, which provides tools for processing images 
+from instruments modeled using the Multi-Slit Spectrograph (MSS) Designer. The library includes functions 
+applying image transformations to map detector images onto the focal plane of the instrument, as well as
+remove diffraction slit curvature from the images. This library is designed to be used stand-alone, using
+the instrument configuration files generated by the MSS Designer. This library is also used internally by
+the MSS Designer.
+
+# Installation
+You can install the library using pip:
+```bash
+pip install mss-imageproc
+```
+
+# Usage
+The library is used in the following way:
+```python
+# Import the library
+from mss_imageproc import MosaicImageStraightener
+
+# Create a straightener object using the instrument configuration file
+straightener = MosaicImageStraightener.from_instrument_config('path/to/instrument_curve_maps.bin')
+# Load an image and map it onto the mosaic grid
+image_array = ...  # Load your image as a 2D NumPy array
+mapped_image = straightener.load_image(image_array)
+# Straighten the image by removing slit curvature
+straightened_images = straightener.straighten_image(mapped_image)
+# The result is a dictionary of straightened images, one for each window. You can access them like this:
+for window_name, straightened_image in straightened_images.items():
+    print(f"Straightened image for window {window_name}:")
+    straightened_image.show()  # Show the straightened image
+```

mss_imageproc-0.0.1/src/mss_imageproc.egg-info/SOURCES.txt

@@ -0,0 +1,9 @@
+README.md
+pyproject.toml
+src/mss_imageproc/__init__.py
+src/mss_imageproc/straighten_image.py
+src/mss_imageproc.egg-info/PKG-INFO
+src/mss_imageproc.egg-info/SOURCES.txt
+src/mss_imageproc.egg-info/dependency_links.txt
+src/mss_imageproc.egg-info/requires.txt
+src/mss_imageproc.egg-info/top_level.txt

mss_imageproc-0.0.1/src/mss_imageproc.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

mss_imageproc-0.0.1/src/mss_imageproc.egg-info/requires.txt

@@ -0,0 +1,10 @@
+astropy>=7.1
+dacite>=1.8.1
+natsort>=8.4.0
+numpy>=1.26
+scikit-image>=0.22
+serde-dataclass>=0.0.4
+tomlkit>=0.12.5
+xarray>=2024.1.0
+netcdf4>=1.6.3
+astropy-xarray==0.1.0

mss_imageproc-0.0.1/src/mss_imageproc.egg-info/top_level.txt

@@ -0,0 +1 @@
+mss_imageproc