As a dev, I want to support PDF files as input to the pipeline in addition to images

[k-allagbe]

Description

Context

Currently, the Fertiscan pipeline only supports image files as input for processing. However, most of the labels we have are in PDF format. To enhance flexibility and usability, we want to extend the pipeline to accept PDF files in addition to images.

Problem Statement

We need to modify the Fertiscan pipeline to allow PDF files as input, extracting images from the PDFs and processing them in the same way as standalone image files.

Acceptance Criteria

• The pipeline can accept PDF files as input in addition to image files.
• PDF files are processed by extracting embedded images and passing them through the same processing steps as standalone image files.
• Appropriate error handling is in place for cases where a PDF does not contain images or has unsupported formats.
• Update documentation to reflect support for PDF input and any limitations.