Tutorials & Usage Examples

Welcome to the Usage Examples section, where we move from theory to practice. Here, you will find a series of hands-on tutorials designed to walk you through the core functionalities of CulicidaeLab. Each example is self-contained, executable, and builds upon the last, demonstrating a complete and logical workflow for mosquito image analysis.

We will follow a common real-world scenario, taking an image and processing it from start to finish. You will learn how to:

Understand and Use the settings Module: We'll start with the foundation. You will learn how to access configurations and understand how the library manages resources. This is the central nervous system of CulicidaeLab.
Managing Datasets: This guide shows you how to use the DatasetsManager to discover, load, and cache the datasets configured within the library. It is essential for anyone looking to perform large-scale model evaluation or exploratory data analysis (EDA).
Detect Mosquitoes: Next, we'll use the MosquitoDetector to answer the first critical question: "Is there a mosquito in this image, and where is it?"
Segment Mosquitoes: For the most detailed analysis, we'll use the MosquitoSegmenter to generate a precise, pixel-level mask of the mosquito's exact shape.
Classify Mosquito Species: Once a mosquito is found, we'll use the MosquitoClassifier to identify its species, a crucial step for epidemiological studies.

Before diving into the examples, please ensure you have successfully installed CulicidaeLab. If you haven't, please see the Installation Guide.

The examples use sample images that are assumed to be in a local test_imgs/ directory. You can download these from our GitHub repository or simply replace the file paths with your own images.

We encourage you to run the code snippets yourself in a Jupyter Notebook or a Python script to get a feel for how everything works together. Let's get started