Image Dataset Analyzer

Build an image dataset analyzer that discovers and reports statistics about images in a directory structure.

Requirements

Your analyzer should:

1. Accept a directory path as input
2. Find all image files recursively within that directory and its subdirectories
3. Count the total number of images found
4. Optionally filter images by a substring in their filename
5. Report the file paths of all discovered images

The analyzer should support common image formats including JPEG, PNG, BMP, and TIFF files.

Implementation

@generates

API

def analyze_images(directory_path, filter_substring=None):
    """
    Analyzes images in a directory structure.

    Args:
        directory_path: Path to the directory to search
        filter_substring: Optional substring to filter filenames

    Returns:
        A dictionary containing:
        - 'count': Total number of images found
        - 'paths': List of image file paths
    """
    pass

Test Cases

• Given a directory with 5 image files (2 JPEG, 2 PNG, 1 BMP) across multiple subdirectories, analyze_images returns count of 5 and all file paths @test
• Given a directory with 10 images where 3 filenames contain "dog", analyze_images with filter_substring="dog" returns count of 3 and only those 3 paths @test
• Given an empty directory, analyze_images returns count of 0 and empty paths list @test

Dependencies { .dependencies }

imutils { .dependency }

Provides image processing utilities including file discovery functions.