DEV Community

Derek
Derek

Posted on

Classifying Architectural Drawings Using PDF SDK with Page Splitting and Data Extraction

An architectural company specializing in design and construction classifies a large number of multi-page architectural PDF drawings to improve collaboration and operational efficiency.

Such companies often undertake numerous architectural design projects, each containing multiple design drawings, such as detailed structural plans, electrical and plumbing plans, mechanical plans, etc. If project drawings can be automatically classified upon upload and provided with an online preview interface for collaborators to review and annotate, it would significantly reduce manual work and greatly streamline the design and review process.

source post: architectural drawing management

Image description

Key Information Extraction - 100% Extraction Accuracy

Each page of an architectural drawing contains key fields - Sheet Number and Title. We can use the data extraction function of the PDF SDK to extract these numbers and titles as the basis for classification. However, the complexity of construction drawing content and elements interferes with the algorithm, leading to incomplete PDF data extraction. ComPDFKit effectively solves this issue: first, the design drawing is split into single pages, and then OCR is performed to extract all Sheet Numbers and Titles from each page.

Each page follows a consistent layout, divided into two sections. The left section contains the design drawing, with specific content varying by project, while the right section consistently displays the basic drawing information. The two key fields (Sheet Number and Title) are located in the fixed right section, maintaining the same position on all pages. Therefore, ComPDFKit identifies each single page and crops its fixed right section for further recognition and extraction. Using this cropping extraction method, a 100% extraction accuracy rate was achieved.

Image description

Frontend Interface Implementation

By integrating the PDF viewing and annotation functions of the Web PDF SDK into the system project, users can easily zoom in and out to inspect details and switch between viewing modes.

With the full set of PDF annotation tools integrated, users can conveniently mark up designs - highlight key points, draw shapes, insert images and links, etc. This greatly enhances collaboration among colleagues.

More Features

Image description

Billboard image

Deploy and scale your apps on AWS and GCP with a world class developer experience

Coherence makes it easy to set up and maintain cloud infrastructure. Harness the extensibility, compliance and cost efficiency of the cloud.

Learn more

Top comments (0)

A Workflow Copilot. Tailored to You.

Pieces.app image

Our desktop app, with its intelligent copilot, streamlines coding by generating snippets, extracting code from screenshots, and accelerating problem-solving.

Read the docs