Building Object Detection Using AI Scan Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional object detection methods in buildings are time-consuming and prone to errors due to manual visual identification, which is inefficient and unreliable.

Innovation Solution

A computer-implemented method using artificial intelligence modules trained with labeled scan data to automatically detect and classify objects within a building, utilizing various scanners to generate data like image, point cloud, radar, and acoustic data, and verifying classifications based on installation rules and georeferenced data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual visual identification by operator is used, then object detection can be performed, but detection time is excessive and error rate is high

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the mechanical human visual identification process with an automated optical scanning system coupled with AI classification algorithms. The scanner captures images of objects, and the AI module automatically classifies them, eliminating manual intervention and significantly reducing detection time while improving accuracy and consistency.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service object detection and classification without requiring human operators. The AI module independently processes scanned images, identifies objects, and generates classification results autonomously, making the detection process self-sufficient and highly efficient.

Inventive Principle:
Principle #25Self-service

2Reliability

If conventional scanning methods are used, then object detection is possible, but operator oversight leads to missed objects

Engineering Contradiction:
Improvedetection completenessVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces the fallible human operator with an automated AI-based detection system that processes all scanned objects systematically. The AI module evaluates each object without fatigue or distraction, ensuring complete detection and classification while reducing the complexity of human training and supervision requirements.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If AI-based automatic detection is implemented, then detection speed and reliability improve, but system complexity increases

Engineering Contradiction:
Improvedetection efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements a universal AI classification module that can handle multiple types of objects and scanning data formats through a single system. The AI module is designed to be versatile, processing various object categories and integrating with different scanner types, which manages system complexity while maintaining high detection efficiency and productivity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3680813B1Method and system for detecting objects installed within a building
Publication Date: 2025.12.24 SIEMENS SCHWEIZ AG
  • EP3680813B1 patent drawingFigure 1~2
  • EP3680813B1 patent drawingFigure 3~4

AI summary

A method and system for detecting objects installed in a building, the system comprising at least one scanner (2) adapted to scan a space within the building to generate scan data (SD) of the respective scanned space; and at least one artificial intelligence module, AIM, (3) implemented on a processing unit (5) of said system (1) adapted to process scan data (SD) generated by said scanner (2) to detect and classify automatically objects installed within the scanned space, wherein the artificial intelligence module, AIM, (3) is trained with labeled training data comprising labels derived from scan data of previously scanned objects provided with associated object identifiers (OID).