Treadscan documentation

_images/treadscan.jpg

Required Python version: python3.8 or higher.

Required packages: numpy, opencv-python, opencv-contrib-python, improutils.

To build this documentation, you need Sphinx, numpydoc and sphinx-rtd-theme.

Treadscan modules:

treadscan.detector

This module is used for detecting stopped vehicle(s) from camera footage.

treadscan.segmentor

This module is used for image segmentation, in particular to find the ellipse defining the tire position in image.

treadscan.extractor

This module is used for tread extraction (unwrapping) given the image and the ellipse defining the vehicle's rim in image.

treadscan.utilities

This module contains various useful methods that can be used in multiple different places.

Treadscan classes:

treadscan.detector.InputType(value)

Enum class specifying types of input for Detector class.

treadscan.detector.FrameExtractor(...)

Class for extracting GRAYSCALE frames from video or folder of images.

treadscan.detector.BackgroundSubtractorSimple(...)

Simple background subtractor immune to sleeping person phenomenon (an object that stops moving won't become a part of the background).

treadscan.detector.Detector(backsub, ...)

Detects presence and motion of a vehicle from footage, yielding 1 image per 1 stopped vehicle.

treadscan.segmentor.SegmentorRCNN(...[, ...])

Uses region based convolutional neural network model to find car wheels in images.

treadscan.segmentor.Segmentor(image)

Contains methods for image segmentation and ellipse detection.

treadscan.extractor.TireModel(image_size)

Model representing a tire.

treadscan.utilities.Ellipse(cx, cy, width, ...)

Class which defines an ellipse.

Indices and tables