Inventory and damage assessment with KaI

Anyone who surveys buildings or documents damage knows how time-consuming, error-prone, and incomplete this process often is. AI can help, as part 12 of the AI ​​series shows.

This article is part of the special topic Artificial Intelligence in Crafts – an assistant called KaI

The detailed recording des current state of a building or roomCreating an accurate data foundation for planning projects like renovations, refurbishments, or conversions is one of the biggest time-wasters. Whether it's the roof, heating system, facade, or interiors: every project begins with a thorough assessment of the current condition, which must be carried out by a qualified professional. Depending on the project, this means several hours of work on-site and more in the office. Our digital assistant, KaI, can significantly reduce this effort, ensures better quality – even if an apprentice is conducting the assessment – ​​and best of all: a virtual view of the property is possible at any time.

There are various possibilities, the Capture to be carried out. This involves the use of different Hardware options. It could be a Smartphone with the appropriate application (App) trade, a 360 degree camera or a 3D laser scanner .

When using a 3D laser scanner, all surfaces are captured line by line or in a raster pattern with a laser beam. This creates millions of measurement points, generating a so-called point cloud. While the human eye can already discern some details in this point cloud, for digital use, the data must first be analyzed and processed to make it suitable for planning in CAD systems. KaI structures this data, adds additional information if necessary, and creates a 3D model.

Simple building structures can be captured using a smartphone and its built-in LiDAR sensor (LiDAR stands for "Light Detection and Ranging" and describes a form of 3D laser scanning). LiDAR scanning is an advancement of laser scanning. It uses invisible laser beams to measure distances. With a 360-degree camera, integrated infrared rangefinders create a spherical panorama from numerous images. This allows distances and proportions within a space to be captured and precisely measured.

The differences between the various solutions lie in the type of technology and the measurement accuracy. Artificial Intelligence works in the background and does it, depending on the application, completely different tasks:

• Data processing 
• Detection of surfaces, slopes, obstacles, ridges, skylights, etc., entirely without manual rework.
• Recognition of building components such as walls, doors, pipes, conduits and conversion of these into elements
• Optimization of the 3D models
• Creating a digital twin
• Locating the images on the corresponding 2D plan
• Easily evaluate data collection. 

Plumbing, heating, and air conditioning technicians benefit: Valves, pipework, and radiators are detected by KaI and located on the plan. This creates a so-called digital twin, which is not only a perfect basis for renovation proposals.

Another field is the Material detectionKaI can distinguish surfaces based on color, texture, and pattern – for example, types of wood, metal, concrete, or plaster. It can also assess conditions, such as whether a surface is freshly painted, dirty, or damaged. KaI also detects corrosion on pipes or fittings and can estimate its severity. In addition to inventory management, damage documentation is therefore a perfect application.

AI series and application examples All parts of the AI ​​series can be found in our special issue "Artificial Intelligence in Crafts – an assistant called KaIFurthermore, handwerksblatt.de offers additional articles on AI, such as the online articles "A mighty clever way to communicate" about the voicemail app with AI chatbot from the start-up "meiti", "AI – how craftsmen benefit in their everyday business"Or"ChatGPT: Save time and money with AI" from the theme special "Digital Crafts". 

From digital image to diagnosis with KaiC

Whether it's water damage, cracking, mold, or corrosion – visual assessment requires expertise and experience. KaI can be used not only in medicine but also in the trades. Here, too, KaI, equipped with image analysis models, was trained using comparison images and learned to reliably identify types of damage. These include cracks in plaster or masonry surfaces, moisture damage, mold growth, corrosion, discoloration, or defective or missing roof tiles.

Kai can

• Identify the type, location, and extent of the damage
• create a damage report
• Make suggestions for possible measures
• estimate the costs based on comparative values ​​and stored price lists.
• Classify the damage according to urgency and create a priority list for repairs.

The success of KaI hinges on data quality. The more a company digitizes and evaluates objects, the more precise KaI becomes, because KaI learns with every data entry. And only if you ensure that KaI learns in a secure environment will the knowledge remain with you. Always remember to protect your data. We'll continue with more inspiration in the next issue.

The author is an employee of the Mittelstand-Digital Zentrum Handwerk (SME Digital Center for Crafts), funded by the Federal Ministry for Economic Affairs and Energy. Since 2016, it has been supporting craft businesses and craft organizations in leveraging the opportunities offered by digital technologies, processes, and business models – free of charge, vendor-neutral, and throughout Germany. Since 2024, it has placed a special focus on artificial intelligence.

handwerkdigital.de

DHB now also digital!Simply click here and register for the digital German Crafts Journal (DHB)!