Max 60% reduction in inspection labor | Automated drawing analysis
Drawing-AI is an AI analysis service for drawings developed by Feature Corporation. By simply importing PDF, paper drawings, or CAD data, the AI interprets the structure within the drawings and automatically processes tasks such as drawing checks, difference verification, and format conversion. It addresses on-site challenges such as "time-consuming drawing checks," "rework due to missed confirmations," and "only specific personnel can read the drawings," transforming the drawings themselves into reusable digital assets. The service covers a wide range of drawings, including circuit diagrams, architectural drawings, and mold drawings, equipped with specialized functions tailored to various applications. For circuit diagrams, it supports structuring bill of materials, exporting to CAD formats, highlighting differences before and after changes, and searching circuits in natural language. For architectural drawings, it allows automatic measurement of dimensions and areas, as well as automatic aggregation of repetitive parts like windows and doors. For mold drawings, it automatically identifies deficiencies or contradictions in dimensions, chamfers, and notes. Thanks to its unique OCR technology, it can also read handwritten drawings and aged blueprints. Outputs are compatible with HTML, PDF, CSV, and CAD formats, and integration with core systems via RPA is possible. For companies that find cloud usage difficult, on-premises implementation is also available.
Inquire About This Product
basic information
<Support Purpose> Optimization of drawing review / Improvement of design quality / Digital asset creation / Technological asset creation (Design support AI) <Reading Target Areas> Electronic circuit diagrams / Architectural drawings / Mold drawings / Equipment wiring diagrams / Electrical system diagrams / Component design drawings / Floor plans / etc. <Functions and Specifications> - Understanding the structure of drawings / Calculation of quantities and estimated costs - Detection of design errors / Listing - Drawing search in natural language - Support for PDF/CAD import (CAD file output) - On-premises support / High security <Implementation Effects> Reduction of design man-hours / Stabilization of quality <Main Functions> - Search - Conversion (circuit diagrams only: PDF → edf) - Comparison - Analysis Automatic analysis and illustration of component elements and connection relationships AI-based element explanation and summarization Drawing review support through automatic alerts for dimensional differences and omissions Output: CAD/CSV/JSON/HTML, etc.
Price information
Please contact us.
Delivery Time
Applications/Examples of results
<Results: Averages> - Work hours Reduced by 30-60% - Work duration Shortened by 1-2 months - Work person-months 4 person-months → 2 person-months <Circuit Diagram> Issue: Inefficiency in circuit redesign due to reading errors and reliance on individuals. Change: Analyzed overall structure and elements. Utilized data digitization and drawing management for circuit diagrams. Enabled pinpoint search of drawings using natural language, allowing even beginners to understand what the drawings represent. Optimized costs through improved work accuracy and shortened work duration. <Architectural Drawings> Issue: Manually counting and listing estimation targets from residential floor plans and store drawings, leading to significant verification work due to dimension omissions. Change: Extracted quantities within equipment, transitioning estimation, procurement, and measurement from visual inspection to AI, reducing work time and minimizing design errors. <Mold Drawings> Issue: Discrepancies between dimensions set in CAD and processing instruction dimensions noted in the drawings, requiring significant time for verification. Change: Detected missing dimensions, errors, and omissions in notes, resulting in a substantial reduction in the labor required for verification and improved efficiency in drawing management. Automated output and transcription to each standard document.
Detailed information
catalog（1）
Download All Catalogs
Company information
Our company started its image recognition business in 2012. Since then, we have leveraged our extensive experience in the fields of computer vision and machine learning to provide low-cost image recognition software for in-vehicle cameras and dashcams. Currently, we are committed to creating fast and high-performance image recognition solutions for "edge intelligence," and we are continuously developing deep learning image recognition technology that is easy to implement on various devices.
