zensvi.cv

Classes

ClassifierGlare

A classifier for identifying glare in images using the model from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierLighting

A classifier for identifying lighting. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierPanorama

A classifier for identifying if an image is a panorama or not. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierPerception

A classifier for evaluating the perception of streetscape based on a given study.

ClassifierPerceptionViT

A classifier for evaluating the perception of streetscape based on a given study

ClassifierPlaces365

A classifier for identifying places using the Places365 model. The model is from Zhou et al. (2017) (https://github.com/CSAILVision/places365).

ClassifierPlatform

A classifier for identifying platform. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierQuality

A classifier for identifying quality. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierReflection

A classifier for identifying reflection. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierViewDirection

A classifier for identifying view_direction. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

ClassifierWeather

A classifier for identifying weather. The model is from Hou et al (2024) (https://github.com/ualsg/global-streetscapes).

DepthEstimator

A class for estimating depth in images.

Embeddings

A class for extracting image embeddings using pre-trained models.

ObjectDetector

Class for detecting objects in images using GroundingDINO model.

Segmenter

A class for performing semantic and panoptic segmentation on images.

Functions

get_low_level_features(→ None)

Processes images from the specified input directory or single image file to

Package Contents

zensvi.cv.get_low_level_features(dir_input: str | pathlib.Path, dir_image_output: str | pathlib.Path = None, dir_summary_output: str | pathlib.Path = None, save_format: str = 'json csv', csv_format: str = 'long', verbosity: int = 1) → None[source]

Processes images from the specified input directory or single image file to detect various low-level features, which include edge detection, blob detection, blur detection, and HSL color space analysis. It optionally saves the processed images and a summary of the features detected.

Parameters:

  • dir_input (Union[str, Path]) – The input directory or image file path.

  • dir_image_output (Union[str, Path], optional) – Directory to save processed images. Defaults to None.

  • dir_summary_output (Union[str, Path], optional) – Directory to save summary results. Defaults to None.

  • save_format (str, optional) – Format to save the summary results. Defaults to “json csv”.

  • csv_format (str, optional) – Format for CSV output. Defaults to “long”.

  • verbosity (int, optional) – Level of verbosity for progress bars. Defaults to 1. 0 = no progress bars, 1 = outer loops only, 2 = all loops.

Returns:

The function does not return any value but outputs results to the specified directories.

Return type:

None

Raises:

ValueError – If neither dir_image_output nor dir_summary_output is provided, indicating that at least one output directory must be specified.