Introduction
The brightest-path-lib
is a Python library which allows users to efficiently find the path with the maximum brightness in a 2D or 3D images. It uses the A* Search and NBA* Search algorithms, which are informed search algorithms that use heuristics to guide the search towards the most promising areas of the image.
Examples
In these examples, a search is performed from the user specified start (green) and stop (red) points. The progress of the search is animated as cyan fill color and the final brightest-path is shown in yellow.
Finding Brightest Path using A* Search
Finding Brightest Path using NBA* Search
Capabilities
With its efficient implementation and intuitive API, this library is a valuable tool for anyone working with 2D or 3D images who needs to identify the path with the maximum brightness.
- The library provides easy-to-use functions that take the image data and start and end points as input and return the path with the maximum brightness as output.
- It supports both grayscale and color images and can handle images of arbitrary sizes.
- The library also provides support for users to know which points in the image are being considered for the brightest path in real-time so that they display them on the original image.
Napi Plugin
We have also created a Napari Tracing Plugin to provide an intuititive graphical-user-interface that uses the `brightest-path-lib`` package. We are using this to trace neuronal dendrites and axons.