"""Standard test images. For more images, see - http://sipi.usc.edu/database/database.php """ import sys from warnings import warn import numpy as np import shutil from ..util.dtype import img_as_bool from ._binary_blobs import binary_blobs from ._registry import registry, legacy_registry, registry_urls from .. import __version__ import os.path as osp import os __all__ = ['data_dir', 'load', 'download_all', 'astronaut', 'binary_blobs', 'brick', 'camera', 'cell', 'checkerboard', 'chelsea', 'clock', 'coffee', 'coins', 'colorwheel', 'grass', 'gravel', 'horse', 'hubble_deep_field', 'immunohistochemistry', 'lbp_frontal_face_cascade_filename', 'lfw_subset', 'logo', 'microaneurysms', 'moon', 'page', 'text', 'retina', 'rocket', 'rough_wall', 'shepp_logan_phantom', 'stereo_motorcycle'] legacy_data_dir = osp.abspath(osp.dirname(__file__)) skimage_distribution_dir = osp.join(legacy_data_dir, '..') try: from pooch.utils import file_hash except ModuleNotFoundError: # Function taken from # https://github.com/fatiando/pooch/blob/master/pooch/utils.py def file_hash(fname, alg="sha256"): """ Calculate the hash of a given file. Useful for checking if a file has changed or been corrupted. Parameters ---------- fname : str The name of the file. alg : str The type of the hashing algorithm Returns ------- hash : str The hash of the file. Examples -------- >>> fname = "test-file-for-hash.txt" >>> with open(fname, "w") as f: ... __ = f.write("content of the file") >>> print(file_hash(fname)) 0fc74468e6a9a829f103d069aeb2bb4f8646bad58bf146bb0e3379b759ec4a00 >>> import os >>> os.remove(fname) """ import hashlib if alg not in hashlib.algorithms_available: raise ValueError( "Algorithm '{}' not available in hashlib".format(alg)) # Calculate the hash in chunks to avoid overloading the memory chunksize = 65536 hasher = hashlib.new(alg) with open(fname, "rb") as fin: buff = fin.read(chunksize) while buff: hasher.update(buff) buff = fin.read(chunksize) return hasher.hexdigest() def _has_hash(path, expected_hash): """Check if the provided path has the expected hash.""" if not osp.exists(path): return False return file_hash(path) == expected_hash def create_image_fetcher(): try: import pooch except ImportError: # Without pooch, fallback on the standard data directory # which for now, includes a few limited data samples return None, legacy_data_dir # Pooch expects a `+` to exist in development versions. # Since scikit-image doesn't follow that convention, we have to manually # remove `.dev` with a `+` if it exists. # This helps pooch understand that it should look in master # to find the required files pooch_version = __version__.replace('.dev', '+') url = "https://github.com/scikit-image/scikit-image/raw/v0.17.x/skimage/" # Create a new friend to manage your sample data storage image_fetcher = pooch.create( # Pooch uses appdirs to select an appropriate directory for the cache # on each platform. # https://github.com/ActiveState/appdirs # On linux this converges to # '$HOME/.cache/scikit-image' # With a version qualifier path=pooch.os_cache("scikit-image"), base_url=url, version=pooch_version, env="SKIMAGE_DATADIR", registry=registry, urls=registry_urls, ) data_dir = osp.join(str(image_fetcher.abspath), 'data') return image_fetcher, data_dir image_fetcher, data_dir = create_image_fetcher() if image_fetcher is None: has_pooch = False else: has_pooch = True def _fetch(data_filename): """Fetch a given data file from either the local cache or the repository. This function provides the path location of the data file given its name in the scikit-image repository. Parameters ---------- data_filename: Name of the file in the scikit-image repository. e.g. 'restoration/tess/camera_rl.npy'. Returns ------- Path of the local file as a python string. Raises ------ KeyError: If the filename is not known to the scikit-image distribution. ModuleNotFoundError: If the filename is known to the scikit-image distribution but pooch is not installed. ConnectionError: If scikit-image is unable to connect to the internet but the dataset has not been downloaded yet. """ resolved_path = osp.join(data_dir, '..', data_filename) expected_hash = registry[data_filename] # Case 1: # The file may already be in the data_dir. # We may have decided to ship it in the scikit-image distribution. if _has_hash(resolved_path, expected_hash): # Nothing to be done, file is where it is expected to be return resolved_path # Case 2: # The user is using a cloned version of the github repo, which # contains both the publicly shipped data, and test data. # In this case, the file would be located relative to the # skimage_distribution_dir gh_repository_path = osp.join(skimage_distribution_dir, data_filename) if _has_hash(gh_repository_path, expected_hash): parent = osp.dirname(resolved_path) os.makedirs(parent, exist_ok=True) shutil.copy2(gh_repository_path, resolved_path) return resolved_path # Case 3: # Pooch not found. if image_fetcher is None: raise ModuleNotFoundError( "The requested file is part of the scikit-image distribution, " "but requires the installation of an optional dependency, pooch. " "To install pooch, use your preferred python package manager. " "Follow installation instruction found at " "https://scikit-image.org/docs/stable/install.html" ) # Case 4: # Pooch needs to download the data. Let the image fetcher to search for # our data. A ConnectionError is raised if no internet connection is # available. try: resolved_path = image_fetcher.fetch(data_filename) except ConnectionError as err: # If we decide in the future to suppress the underlying 'requests' # error, change this to `raise ... from None`. See PEP 3134. raise ConnectionError( 'Tried to download a scikit-image dataset, but no internet ' 'connection is available. To avoid this message in the ' 'future, try `skimage.data.download_all()` when you are ' 'connected to the internet.' ) from err return resolved_path def _init_pooch(): os.makedirs(data_dir, exist_ok=True) shutil.copy2(osp.join(skimage_distribution_dir, 'data', 'README.txt'), osp.join(data_dir, 'README.txt')) data_base_dir = osp.join(data_dir, '..') # Fetch all legacy data so that it is available by default for filename in legacy_registry: _fetch(filename) # This function creates directories, and has been the source of issues for # downstream users, see # https://github.com/scikit-image/scikit-image/issues/4660 # https://github.com/scikit-image/scikit-image/issues/4664 if has_pooch: _init_pooch() def download_all(directory=None): """Download all datasets for use with scikit-image offline. Scikit-image datasets are no longer shipped with the library by default. This allows us to use higher quality datasets, while keeping the library download size small. This function requires the installation of an optional dependency, pooch, to download the full dataset. Follow installation instruction found at https://scikit-image.org/docs/stable/install.html Call this function to download all sample images making them available offline on your machine. Parameters ---------- directory: path-like, optional The directory where the dataset should be stored. Raises ------ ModuleNotFoundError: If pooch is not install, this error will be raised. Notes ----- scikit-image will only search for images stored in the default directory. Only specify the directory if you wish to download the images to your own folder for a particular reason. You can access the location of the default data directory by inspecting the variable `skimage.data.data_dir`. """ if image_fetcher is None: raise ModuleNotFoundError( "To download all package data, scikit-image needs an optional " "dependency, pooch." "To install pooch, follow our installation instructions found at " "https://scikit-image.org/docs/stable/install.html" ) # Consider moving this kind of logic to Pooch old_dir = image_fetcher.path try: if directory is not None: image_fetcher.path = directory for filename in image_fetcher.registry: _fetch(filename) finally: image_fetcher.path = old_dir def lbp_frontal_face_cascade_filename(): """Return the path to the XML file containing the weak classifier cascade. These classifiers were trained using LBP features. The file is part of the OpenCV repository [1]_. References ---------- .. [1] OpenCV lbpcascade trained files https://github.com/opencv/opencv/tree/master/data/lbpcascades """ return _fetch('data/lbpcascade_frontalface_opencv.xml') def load(f, as_gray=False): """Load an image file located in the data directory. Parameters ---------- f : string File name. as_gray : bool, optional Whether to convert the image to grayscale. Returns ------- img : ndarray Image loaded from ``skimage.data_dir``. Notes ----- This functions is deprecated and will be removed in 0.18. """ warn('This function is deprecated and will be removed in 0.18. ' 'Use `skimage.io.load` or `imageio.imread` directly.', stacklevel=2) return _load(f, as_gray=as_gray) def _load(f, as_gray=False): """Load an image file located in the data directory. Parameters ---------- f : string File name. as_gray : bool, optional Whether to convert the image to grayscale. Returns ------- img : ndarray Image loaded from ``skimage.data_dir``. """ # importing io is quite slow since it scans all the backends # we lazy import it here from ..io import imread return imread(_fetch(f), as_gray=as_gray) def camera(): """Gray-level "camera" image. Often used for segmentation and denoising examples. Returns ------- camera : (512, 512) uint8 ndarray Camera image. """ return _load("data/camera.png") def astronaut(): """Color image of the astronaut Eileen Collins. Photograph of Eileen Collins, an American astronaut. She was selected as an astronaut in 1992 and first piloted the space shuttle STS-63 in 1995. She retired in 2006 after spending a total of 38 days, 8 hours and 10 minutes in outer space. This image was downloaded from the NASA Great Images database `__. No known copyright restrictions, released into the public domain. Returns ------- astronaut : (512, 512, 3) uint8 ndarray Astronaut image. """ return _load("data/astronaut.png") def brick(): """Brick wall. Returns ------- brick : (512, 512) uint8 image A small section of a brick wall. Notes ----- The original image was downloaded from `CC0Textures `_ and licensed under the Creative Commons CC0 License. A perspective transform was then applied to the image, prior to rotating it by 90 degrees, cropping and scaling it to obtain the final image. """ """ The following code was used to obtain the final image. >>> import sys; print(sys.version) >>> import platform; print(platform.platform()) >>> import skimage; print(f"scikit-image version: {skimage.__version__}") >>> import numpy; print(f"numpy version: {numpy.__version__}") >>> import imageio; print(f"imageio version {imageio.__version__}") 3.7.3 | packaged by conda-forge | (default, Jul 1 2019, 21:52:21) [GCC 7.3.0] Linux-5.0.0-20-generic-x86_64-with-debian-buster-sid scikit-image version: 0.16.dev0 numpy version: 1.16.4 imageio version 2.4.1 >>> import requests >>> import zipfile >>> url = 'https://cdn.struffelproductions.com/file/cc0textures/Bricks25/%5B2K%5DBricks25.zip' >>> r = requests.get(url) >>> with open('[2K]Bricks25.zip', 'bw') as f: ... f.write(r.content) >>> with zipfile.ZipFile('[2K]Bricks25.zip') as z: ... z.extract('Bricks25_col.jpg') >>> from numpy.linalg import inv >>> from skimage.transform import rescale, warp, rotate >>> from skimage.color import rgb2gray >>> from imageio import imread, imwrite >>> from skimage import img_as_ubyte >>> import numpy as np >>> # Obtained playing around with GIMP 2.10 with their perspective tool >>> H = inv(np.asarray([[ 0.54764, -0.00219, 0], ... [-0.12822, 0.54688, 0], ... [-0.00022, 0, 1]])) >>> brick_orig = imread('Bricks25_col.jpg') >>> brick = warp(brick_orig, H) >>> brick = rescale(brick[:1024, :1024], (0.5, 0.5, 1)) >>> brick = rotate(brick, -90) >>> imwrite('brick.png', img_as_ubyte(rgb2gray(brick))) """ return _load("data/brick.png", as_gray=True) def grass(): """Grass. Returns ------- grass : (512, 512) uint8 image Some grass. Notes ----- The original image was downloaded from `DeviantArt `__ and licensed underthe Creative Commons CC0 License. The downloaded image was cropped to include a region of ``(512, 512)`` pixels around the top left corner, converted to grayscale, then to uint8 prior to saving the result in PNG format. """ """ The following code was used to obtain the final image. >>> import sys; print(sys.version) >>> import platform; print(platform.platform()) >>> import skimage; print(f"scikit-image version: {skimage.__version__}") >>> import numpy; print(f"numpy version: {numpy.__version__}") >>> import imageio; print(f"imageio version {imageio.__version__}") 3.7.3 | packaged by conda-forge | (default, Jul 1 2019, 21:52:21) [GCC 7.3.0] Linux-5.0.0-20-generic-x86_64-with-debian-buster-sid scikit-image version: 0.16.dev0 numpy version: 1.16.4 imageio version 2.4.1 >>> import requests >>> import zipfile >>> url = 'https://images-wixmp-ed30a86b8c4ca887773594c2.wixmp.com/f/a407467e-4ff0-49f1-923f-c9e388e84612/d76wfef-2878b78d-5dce-43f9-be36->> 26ec9bc0df3b.jpg?token=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJzdWIiOiJ1cm46YXBwOjdlMGQxODg5ODIyNjQzNzNhNWYwZDQxNWVhMGQyNmUwIiwiaXNzIjoidXJuOmFwcDo3ZTBkMTg4OTgyMjY0MzczYTVmMGQ0MTVlYTBkMjZlMCIsIm9iaiI6W1t7InBhdGgiOiJcL2ZcL2E0MDc0NjdlLTRmZjAtNDlmMS05MjNmLWM5ZTM4OGU4NDYxMlwvZDc2d2ZlZi0yODc4Yjc4ZC01ZGNlLTQzZjktYmUzNi0yNmVjOWJjMGRmM2IuanBnIn1dXSwiYXVkIjpbInVybjpzZXJ2aWNlOmZpbGUuZG93bmxvYWQiXX0.98hIcOTCqXWQ67Ec5bM5eovKEn2p91mWB3uedH61ynI' >>> r = requests.get(url) >>> with open('grass_orig.jpg', 'bw') as f: ... f.write(r.content) >>> grass_orig = imageio.imread('grass_orig.jpg') >>> grass = skimage.img_as_ubyte(skimage.color.rgb2gray(grass_orig[:512, :512])) >>> imageio.imwrite('grass.png', grass) """ return _load("data/grass.png", as_gray=True) def gravel(): """Gravel Returns ------- gravel : (512, 512) uint8 image Grayscale gravel sample. Notes ----- The original image was downloaded from `CC0Textures `__ and licensed under the Creative Commons CC0 License. The downloaded image was then rescaled to ``(1024, 1024)``, then the top left ``(512, 512)`` pixel region was cropped prior to converting the image to grayscale and uint8 data type. The result was saved using the PNG format. """ """ The following code was used to obtain the final image. >>> import sys; print(sys.version) >>> import platform; print(platform.platform()) >>> import skimage; print(f"scikit-image version: {skimage.__version__}") >>> import numpy; print(f"numpy version: {numpy.__version__}") >>> import imageio; print(f"imageio version {imageio.__version__}") 3.7.3 | packaged by conda-forge | (default, Jul 1 2019, 21:52:21) [GCC 7.3.0] Linux-5.0.0-20-generic-x86_64-with-debian-buster-sid scikit-image version: 0.16.dev0 numpy version: 1.16.4 imageio version 2.4.1 >>> import requests >>> import zipfile >>> url = 'https://cdn.struffelproductions.com/file/cc0textures/Gravel04/%5B2K%5DGravel04.zip' >>> r = requests.get(url) >>> with open('[2K]Gravel04.zip', 'bw') as f: ... f.write(r.content) >>> with zipfile.ZipFile('[2K]Gravel04.zip') as z: ... z.extract('Gravel04_col.jpg') >>> from skimage.transform import resize >>> gravel_orig = imageio.imread('Gravel04_col.jpg') >>> gravel = resize(gravel_orig, (1024, 1024)) >>> gravel = skimage.img_as_ubyte(skimage.color.rgb2gray(gravel[:512, :512])) >>> imageio.imwrite('gravel.png', gravel) """ return _load("data/gravel.png", as_gray=True) def text(): """Gray-level "text" image used for corner detection. Notes ----- This image was downloaded from Wikipedia `__. No known copyright restrictions, released into the public domain. Returns ------- text : (172, 448) uint8 ndarray Text image. """ return _load("data/text.png") def checkerboard(): """Checkerboard image. Checkerboards are often used in image calibration, since the corner-points are easy to locate. Because of the many parallel edges, they also visualise distortions particularly well. Returns ------- checkerboard : (200, 200) uint8 ndarray Checkerboard image. """ return _load("data/chessboard_GRAY.png") def cell(): """Cell floating in saline. This is a quantitative phase image retrieved from a digital hologram using the Python library ``qpformat``. The image shows a cell with high phase value, above the background phase. Because of a banding pattern artifact in the background, this image is a good test of thresholding algorithms. The pixel spacing is 0.107 µm. These data were part of a comparison between several refractive index retrieval techniques for spherical objects as part of [1]_. This image is CC0, dedicated to the public domain. You may copy, modify, or distribute it without asking permission. Returns ------- cell : (660, 550) uint8 array Image of a cell. References ---------- .. [1] Paul Müller, Mirjam Schürmann, Salvatore Girardo, Gheorghe Cojoc, and Jochen Guck. "Accurate evaluation of size and refractive index for spherical objects in quantitative phase imaging." Optics Express 26(8): 10729-10743 (2018). :DOI:`10.1364/OE.26.010729` """ return _load('data/cell.png') def coins(): """Greek coins from Pompeii. This image shows several coins outlined against a gray background. It is especially useful in, e.g. segmentation tests, where individual objects need to be identified against a background. The background shares enough grey levels with the coins that a simple segmentation is not sufficient. Notes ----- This image was downloaded from the `Brooklyn Museum Collection `__. No known copyright restrictions. Returns ------- coins : (303, 384) uint8 ndarray Coins image. """ return _load("data/coins.png") def logo(): """Scikit-image logo, a RGBA image. Returns ------- logo : (500, 500, 4) uint8 ndarray Logo image. """ return _load("data/logo.png") def microaneurysms(): """Gray-level "microaneurysms" image. Detail from an image of the retina (green channel). The image is a crop of image 07_dr.JPG from the High-Resolution Fundus (HRF) Image Database: https://www5.cs.fau.de/research/data/fundus-images/ Notes ----- No copyright restrictions. CC0 given by owner (Andreas Maier). Returns ------- microaneurysms : (102, 102) uint8 ndarray Retina image with lesions. References ---------- .. [1] Budai, A., Bock, R, Maier, A., Hornegger, J., Michelson, G. (2013). Robust Vessel Segmentation in Fundus Images. International Journal of Biomedical Imaging, vol. 2013, 2013. :DOI:`10.1155/2013/154860` """ return _load("data/microaneurysms.png") def moon(): """Surface of the moon. This low-contrast image of the surface of the moon is useful for illustrating histogram equalization and contrast stretching. Returns ------- moon : (512, 512) uint8 ndarray Moon image. """ return _load("data/moon.png") def page(): """Scanned page. This image of printed text is useful for demonstrations requiring uneven background illumination. Returns ------- page : (191, 384) uint8 ndarray Page image. """ return _load("data/page.png") def horse(): """Black and white silhouette of a horse. This image was downloaded from `openclipart ` No copyright restrictions. CC0 given by owner (Andreas Preuss (marauder)). Returns ------- horse : (328, 400) bool ndarray Horse image. """ return img_as_bool(_load("data/horse.png", as_gray=True)) def clock(): """Motion blurred clock. This photograph of a wall clock was taken while moving the camera in an aproximately horizontal direction. It may be used to illustrate inverse filters and deconvolution. Released into the public domain by the photographer (Stefan van der Walt). Returns ------- clock : (300, 400) uint8 ndarray Clock image. """ return _load("data/clock_motion.png") def immunohistochemistry(): """Immunohistochemical (IHC) staining with hematoxylin counterstaining. This picture shows colonic glands where the IHC expression of FHL2 protein is revealed with DAB. Hematoxylin counterstaining is applied to enhance the negative parts of the tissue. This image was acquired at the Center for Microscopy And Molecular Imaging (CMMI). No known copyright restrictions. Returns ------- immunohistochemistry : (512, 512, 3) uint8 ndarray Immunohistochemistry image. """ return _load("data/ihc.png") def chelsea(): """Chelsea the cat. An example with texture, prominent edges in horizontal and diagonal directions, as well as features of differing scales. Notes ----- No copyright restrictions. CC0 by the photographer (Stefan van der Walt). Returns ------- chelsea : (300, 451, 3) uint8 ndarray Chelsea image. """ return _load("data/chelsea.png") def coffee(): """Coffee cup. This photograph is courtesy of Pikolo Espresso Bar. It contains several elliptical shapes as well as varying texture (smooth porcelain to course wood grain). Notes ----- No copyright restrictions. CC0 by the photographer (Rachel Michetti). Returns ------- coffee : (400, 600, 3) uint8 ndarray Coffee image. """ return _load("data/coffee.png") def hubble_deep_field(): """Hubble eXtreme Deep Field. This photograph contains the Hubble Telescope's farthest ever view of the universe. It can be useful as an example for multi-scale detection. Notes ----- This image was downloaded from `HubbleSite `__. The image was captured by NASA and `may be freely used in the public domain `_. Returns ------- hubble_deep_field : (872, 1000, 3) uint8 ndarray Hubble deep field image. """ return _load("data/hubble_deep_field.jpg") def retina(): """Human retina. This image of a retina is useful for demonstrations requiring circular images. Notes ----- This image was downloaded from `wikimedia `. This file is made available under the Creative Commons CC0 1.0 Universal Public Domain Dedication. References ---------- .. [1] Häggström, Mikael (2014). "Medical gallery of Mikael Häggström 2014". WikiJournal of Medicine 1 (2). :DOI:`10.15347/wjm/2014.008`. ISSN 2002-4436. Public Domain Returns ------- retina : (1411, 1411, 3) uint8 ndarray Retina image in RGB. """ return _load("data/retina.jpg") def shepp_logan_phantom(): """Shepp Logan Phantom. References ---------- .. [1] L. A. Shepp and B. F. Logan, "The Fourier reconstruction of a head section," in IEEE Transactions on Nuclear Science, vol. 21, no. 3, pp. 21-43, June 1974. :DOI:`10.1109/TNS.1974.6499235` Returns ------- phantom : (400, 400) float64 image Image of the Shepp-Logan phantom in grayscale. """ return _load("data/phantom.png", as_gray=True) def colorwheel(): """Color Wheel. Returns ------- colorwheel : (370, 371, 3) uint8 image A colorwheel. """ return _load("data/color.png") def rocket(): """Launch photo of DSCOVR on Falcon 9 by SpaceX. This is the launch photo of Falcon 9 carrying DSCOVR lifted off from SpaceX's Launch Complex 40 at Cape Canaveral Air Force Station, FL. Notes ----- This image was downloaded from `SpaceX Photos `__. The image was captured by SpaceX and `released in the public domain `_. Returns ------- rocket : (427, 640, 3) uint8 ndarray Rocket image. """ return _load("data/rocket.jpg") def stereo_motorcycle(): """Rectified stereo image pair with ground-truth disparities. The two images are rectified such that every pixel in the left image has its corresponding pixel on the same scanline in the right image. That means that both images are warped such that they have the same orientation but a horizontal spatial offset (baseline). The ground-truth pixel offset in column direction is specified by the included disparity map. The two images are part of the Middlebury 2014 stereo benchmark. The dataset was created by Nera Nesic, Porter Westling, Xi Wang, York Kitajima, Greg Krathwohl, and Daniel Scharstein at Middlebury College. A detailed description of the acquisition process can be found in [1]_. The images included here are down-sampled versions of the default exposure images in the benchmark. The images are down-sampled by a factor of 4 using the function `skimage.transform.downscale_local_mean`. The calibration data in the following and the included ground-truth disparity map are valid for the down-sampled images:: Focal length: 994.978px Principal point x: 311.193px Principal point y: 254.877px Principal point dx: 31.086px Baseline: 193.001mm Returns ------- img_left : (500, 741, 3) uint8 ndarray Left stereo image. img_right : (500, 741, 3) uint8 ndarray Right stereo image. disp : (500, 741, 3) float ndarray Ground-truth disparity map, where each value describes the offset in column direction between corresponding pixels in the left and the right stereo images. E.g. the corresponding pixel of ``img_left[10, 10 + disp[10, 10]]`` is ``img_right[10, 10]``. NaNs denote pixels in the left image that do not have ground-truth. Notes ----- The original resolution images, images with different exposure and lighting, and ground-truth depth maps can be found at the Middlebury website [2]_. References ---------- .. [1] D. Scharstein, H. Hirschmueller, Y. Kitajima, G. Krathwohl, N. Nesic, X. Wang, and P. Westling. High-resolution stereo datasets with subpixel-accurate ground truth. In German Conference on Pattern Recognition (GCPR 2014), Muenster, Germany, September 2014. .. [2] http://vision.middlebury.edu/stereo/data/scenes2014/ """ filename = _fetch("data/motorcycle_disp.npz") disp = np.load(filename)['arr_0'] return (_load("data/motorcycle_left.png"), _load("data/motorcycle_right.png"), disp) def lfw_subset(): """Subset of data from the LFW dataset. This database is a subset of the LFW database containing: * 100 faces * 100 non-faces The full dataset is available at [2]_. Returns ------- images : (200, 25, 25) uint8 ndarray 100 first images are faces and subsequent 100 are non-faces. Notes ----- The faces were randomly selected from the LFW dataset and the non-faces were extracted from the background of the same dataset. The cropped ROIs have been resized to a 25 x 25 pixels. References ---------- .. [1] Huang, G., Mattar, M., Lee, H., & Learned-Miller, E. G. (2012). Learning to align from scratch. In Advances in Neural Information Processing Systems (pp. 764-772). .. [2] http://vis-www.cs.umass.edu/lfw/ """ return np.load(_fetch('data/lfw_subset.npy'))