Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{188164,
        author = {Deepak T and Kavita Reddy and Ganesh Y V and Felix Xavier J},
        title = {AI BASED BLUEPRINT ERROR DETECTION},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {1380-1389},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=188164},
        abstract = {Persistent errors in technical documentation, which result in expensive rework, safety risks, and project delays, pose serious operational and financial risks to the Architecture, Engineering, and Construction (AEC) sector. [1] This paper suggests an advanced, integrated deep learning framework for automated blueprint quality control as a solution to this problem. The system uses a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) based Optical Character Recognition (OCR) path for the accurate extraction of semantic annotations and dimensions, in conjunction with an optimized single-stage object detection architecture (YOLO) for reliable geometric entity localization. [2] A spatial-semantic Late Multimodal Fusion technique that aligns disparate visual and textual features is the main methodological advancement. A deterministic Rule-Based Constraint Validation Engine then uses this aligned data to transparently verifyA deterministic Rule-Based Constraint Validation Engine then uses this aligned data to transparently check for geometric conflicts, dimensional inconsistencies, and design code compliance. [3] With an overall Macro Average F1Score of 95.5%, a thorough evaluation shows how effective the system is. This balanced performance highlights the model's capacity to reduce both time-consuming false positives and extremely harmful false negatives (missed critical errors), providing a clear and dependable route toward digital transformation in documentation review. [4,5]},
        keywords = {YOLO, Multimodal Fusion, Optical Character Recognition (OCR), Computer Vision, Document Image Analysis, Architectural Engineering, Deep Learning, Design Constraint Checking, and Anomaly Detection.},
        month = {December},
        }