ZEISS Automated Defect Detection (ZADD)

Artificial intelligence (AI) in computed tomography (CT)

Reliable evaluation software for inline and atline X-ray inspection

The ZEISS Automated Defect Detection (ZADD) software detects even small and fuzzy defects in components reliably and quickly. The Automated Defect Detection (ADD) software, also called Automated Defect Recognition (ADR) software, can be used not only for injection molded parts, but also for batteries and medical or additively manufactured components.

ZEISS offers everything from a single source: with the complete ZEISS X-ray portfolio, you can use artificial intelligence (AI) in 3D computed tomography (CT) and 2D X-ray technology both inline and atline. ZEISS Automated Defect Detection (ZADD) software is compatible with all ZEISS computed tomography systems.

Your advantages with the ZEISS Automated Defect Detection software at a glance

A stopwatch symbolizes the efficiency gains that can be achieved with the help of AI in CT.

Defect analysis in only 60 seconds in inline mode

  • Minimization of test cycle times
  • Faster scans/defect detection
  • Time savings for operators
A checklist shows that the Automated Defect Detection software can be used to create clear reports.

Robust quality results & clear reporting

  • Perfect results, even if image quality is not perfect
  • Suitable for mixed and dense materials
The image shows an arrow stuck in the center of a target to show the precise defect detection of the AI software in CT.

No parameter tuning required

  • Subjective decisions are avoided
  • No reference data required in atline mode
A person with a light bulb next to their head is shown to represent the custom optimization of the AI Defect Detection software.

User-defined ZEISS Automated Defect Detection software

  • Custom optimization of defect analysis
  • User-defined programming of the software

ZEISS Automated Defect Detection

AI software for your application areas

ZEISS Automated Defect Detection machine learning software sets new standards by applying AI to 3D CT and 2D X-ray systems with CT option. It detects, localizes and classifies defects or anomalies (irregularities) while analyzing them in detail by reading CT-scans and X-rays. Based on inspection regions, the AI software evaluates whether the defect will pose a problem after further processing steps and whether the component must be rejected as a result. If similar defects occur more frequently in inline operation, this can be detected and thus intervention in the production process can take place at an early stage to reduce scrap and save costs. In atline operation, the ZADD software is convincing in optimizing and monitoring the casting process as well as improved and faster component development.

The image shows a component that can be inspected for defects using artificial intelligence in CT.

Reliably detect defects in components

During the complex manufacturing process of components, different defects can occur. Especially inside the part, defects are invisible to the naked eye and can have a major impact on the stability and functionality of the component. Artificial intelligence combined with industrial computed tomography or 2D X-ray technology makes these hidden problem areas visible, analyzes them and detects them at an early stage. The ZEISS Automated Defect Detection software specializes in the detection of different defects, so that defects can be detected quickly and reliably even in poor image quality with many artifacts.

Inline defect analysis in just 60 seconds

In order to be able to sort out defective components at an early stage in a value chain, the 3D data must be evaluated reliably and quickly during an inline inspection. In just 60 seconds, the ZEISS Automated Defect Detection analysis software reliably examines 4 billion voxels (3D pixels) for defects that are only a few voxels in size. Components with critical defects are thus accurately sorted out or, if possible, subjected to rework. Good parts, on the other hand, pass through the further machining process unhindered. The result: a lower reject rate and high component quality. In this way, you achieve a constant increase in efficiency and maximum process reliability with AI in CT.

The image shows the performance of an inline inspection completed in just 60 seconds using AI in CT.
The image shows the AI software reading a CT scan of a component and detecting a defect.

Reliable evaluation

If the ZEISS ADD software detects a defect, the software evaluates it in terms of position, shape, size and type. For example, if pores are close to a surface that will be machined in the further process, they may pose a greater risk than elsewhere inside the component. The software is able to predict whether the defect may cause problems in subsequent machining steps. The component is then automatically rejected at an early stage. This saves you time and money. In addition, you can define in the ZADD software under which criteria a defect still meets the quality standards and when it is classified as critical. With ZEISS Automated Defect Detection, you get a solution that is tailored precisely to your requirements.

Clear reporting

If the ZEISS Automated Defect Detection software finds a critical defect in inline operation, it creates a report using the ZEISS PiWeb data management software. This gives you the opportunity to view and evaluate defects again afterwards in a 3D view. If your manual inspection leads you to the conclusion that the defect was caused by sand residue, for example, you can easily remove this source of defect and avoid unnecessary scrap. This results in high cost savings. The software also enables results to be displayed in tabular form and output in common formats such as csv.

The image shows the reporting tool of the Automated Defect Detection software.

Typical defects on and in components

The image shows a component with pores scanned by CT AI.


A pore is a spherical or ellipsoidal cavity with mostly smooth walls inside the component. Depending on their origin, they can contain air, vapor, hydrogen, or other gases (e.g. from lubricants). They often occur in upper casting layer, but in poorly evacuated areas or undercuts they can be distributed within the whole casting.

The image shows a component with a cold seal, which was analyzed with artificial intelligence in CT.

Cold run / cold shut

Cold run occurs preferably on flat surfaces with relatively low thickness. This can result in a separation of the cohesion, leaving holes, areas that have not run out, but also rounded edges and overlaps. In die casting, cold run can be seen on very fine and thin surface slates.

A component with a core fracture, which can be detected with the Automated Defect Detection software.

Core breakage

Damage to the casting core, such as breaking off or breaking apart, leads to defective component geometries during the casting process, which can have a significant impact on functionality. This can be caused by suboptimal molding material composition or excessive thermal stress.

Shown is a component with a blowhole that can be identified by an AI reading CT scans.

Shrinkage hole

Shrinkage holes occur through the shrinkage process during the solidification of melted metal. The degree of shrinkage depends on the melting point of the material. As the outer shell and sprue of a casting generally solidify early, volume deficits can be formed in the interior of the casting. Shrinkage is characterized by rough and spongy surfaces and is more elongated.


Microporosity can be detected by an automatic defect detection software.


Micro-porosity can be understood as an accumulation of small shrinkage holes (micro-shrinkage / interdendritic shrinkage), which can create chains and lead to leakages. This porosity appears in a CT scan with lower resolution as spongious areas.

Residual sand can be localized by the AI reading the CT scan.

Sand residual

Through gravity casting and low-pressure die casting, residuals of sand (or salt) cores may remain in the interiors of parts, when the coring process is not sufficient.

 A wall shift which can be identified in the with AI in CT.

Wall displacement

If, for example, defects occur in the positioning of the core in the mold before casting, or if the cores shift during the casting process, the geometries of the casting no longer match the CAD model.

The AI software can detect chips in the CT scan of the component.


During rough machining of the component (e.g. saw cut on the feeder), aluminum chips are produced which can fall into the component. Likewise, small protrusions (feathers) can break off during coring and remain in the component. These aluminum residues can lead to defects in the cooling system, for example, during subsequent operation.

Inclusions, as seen here, are detected by Automated Defect Detection.


Inclusions are partially or completely embedded impurities in the cast component which are usually denser than the base material. They are caused, for example, by foreign bodies in the casting mold or by contaminated casting material.

ZEISS Inline X-Ray Systems with Automated Defect Detection

ZEISS Automated Defect Detection is compatible with any ZEISS computed tomography system and can be easily integrated into your process. During continuous inline measurement along the entire process, ZADD delivers precise results and detects the smallest defects or flaws early on.

ZEISS also offers a complete solution from a single source: with the ZEISS VoluMax inline X-ray system, you have the option of using AI for reading 3D CT scans with perfectly coordinated measurement technology and software.

ZEISS VoluMax inline X-Ray system with AI software ZADD.


  • Fully automated, non-destructive inline inspection
  • Fast inspection of many workpieces in 3D
  • Individual configuration depending on measurement task by ZEISS
  • Choice of manual or automatic loading
  • With Golden Part Inspection

ZEISS Atline X-Ray Systems with Automated Defect Detection

Atline measurements also benefit from AI in CT. Complex measurement procedures can be performed in the quality lab without the environment of production influencing the results. The ZEISS Automated Defect Detection software makes manual measurements more precise and detects defects in production quickly and accurately. In combination with ZEISS atline X-ray systems and GOM Volume Inspect Pro, you can achieve compliance with high inspection standards and efficient quality control.

ZEISS METROTOM atline X-Ray system with AI software ZADD.


  • Complex measurements and inspections
  • Linear guides and rotary table
  • Metrological traceability
  • Small footprint with the ZEISS METROTOM 1
  • Comprehensive computed tomography data analysis in 3D with GOM Volume Inspect Pro
Atline X-ray system ZEISS BOSELLO with AI software ZADD.


  • Industrial 2D X-ray inspection
  • Optimal for high volume production
  • High throughput
  • Flexible application in many industries
  • Atline CT option expandable with AI reading X-rays

Extensive service offer for ZEISS Automated Defect Detection users


The ZEISS team accompanies you from the very beginning! You will receive support in the design and programming of the algorithm, so that your ZEISS X-ray system in combination with the ZEISS Automated Defect Detection software will contribute to process reliability and increased efficiency.


The ZEISS team is available to help you adapt and optimize the algorithm of the ZEISS Automated Defect Detection software. ZEISS is also your competent partner for other questions and applications related to metrology.

Contact us.


ZEISS caters to your application areas and needs! The AI-driven ZEISS Automated Defect Detection software can be trained with an individual model to become a customized solution for your measurement task.

Artificial Intelligence in X-Ray Technology

AI drives CT forward

Artificial intelligence is ubiquitous. Autonomous driving is just one of many examples of the application of AI. Artificial intelligence is also a topic in industry and thus in computed tomography and is becoming increasingly significant. This is because it enables defect analyses to be carried out even more reliably, accurately and quickly. In industry, a defect is often located inside a component. An optical inspection process for quality control is then no longer sufficient because it does not provide any indication of internal defects. X-ray inspection allows a close look inside a component and can thus detect defects at an early stage. By using AI in 3D CT and 2D X-ray inspection, a partially automated defect analysis is realized.


Explanation of terms:
In connection with AI and CT, the terms AI Defect Detection or AI Anomaly Detection are often used. AI stands for Artificial Intelligence and Defect Detection or Anomaly Detection means defect detection or anomaly detection. The addition of "NDT" makes it clear that AI works non-destructively, because NDT stands for non-destructive testing.


How the ZEISS Automated Defect Detection software works

The ZADD software has several features that you can use optimally in either your inline or atline process. In the slider, the features and for which solution they are suitable are explained in detail.

The image shows an anomaly detection by AI in a CT scan.
The image shows a defect detected in a CT scan by AI.
The image shows an anomaly detected in the CT scan by using AI.

Recognition (Where?)

  • The software will detect anomalies regardless of defect type
  • The overall volume for thorough analyses is drastically reduced – the inspection process is decisively accelerated
  • The process is based on classic machine learning methods
    • Enables fast and easy adaptation to new components
    • AI teach-in is completed after only 30-50 scans of good parts
Anomalies are assigned to different defect types by the AI.

Classification (What?)

  • The defect type of each anomaly found is determined
  • Depending on the type of defect, the part is forwarded for rework, toleranced as a good part while still within tolerance, or classified as scrap
  • The process is based on Deep Learning
    • No customer-specific training necessary
    • Parts with uncertain classification can be forwarded to an operator for manual inspection


A CT scan before segmentation of the AI.
A CT scan after segmentation of the AI.

Segmentation (How?)

Inline & Atline
  • ZADD software determines the exact 3D shape of the defect
  • This allows the following to be accurately calculated:
    • Exact defect sizes
    • Defect volume
    • Distance to the machined surface
  • The process is based on Deep Learning
The image shows a suggestion of the Automated Defect Detection software regarding a defect on a component.

Decision (OK or not OK?)

Inline with ZEISS PiWeb
  • The decision whether the part is a good part or not is based on predefined values and tolerances
  • The X-ray inspection tolerance workflow is rule-based and does not need to be re-trained
  • The software is easy to understand and customized for application experts
  • Defined regions allow differentiated decisions regarding relevance
    • That is, in one place the defect may be problematic, in another place it is tolerable
  • The operator can overrule the suggestions of the ZADD software
The image shows the database view of the Automated Defect Detection software.

Process statistics (When?)

  • All defects (sectional views and properties) are collected in a database
  • Archiving of data allows long-term tracking of defects
  • ZADD software provides advanced visualization features for a holistic view
  • Statistical process control enables the detection of process correlations and optimizations
  • Future trend: With the use of machine learning, process deviations can be detected more quickly

When will AI be used in CT?

Artificial intelligence is a trend in automation. Process requirements are becoming tighter and tighter, and even in harsh measurement environments, image evaluation and defect analysis must work quickly and reliably. This is especially true for safety-relevant components, e.g. in the automotive industry or aerospace. To increase quality by performing defect analyses faster, while at the same time offering high process reliability, AI is used for reading CT-scans. Defect detection with AI eliminates the need for manual tuning of parameters, thus avoiding subjective decisions in defect detection.

ZEISS Automated Defect Detection is particularly useful when volume data is affected by overly dense materials or short scan times. While artifacts and noise in the images usually cause faulty detections, the software remains unaffected by these effects.

Take a look at specific examples of how AI is used in 3D CT, but also in X-ray inspection in general:

These industries benefit from AI with ZEISS Automated Defect Detection

New Energy Vehicles (NEV)

From Energy to eMotion with ZEISS eMobility Solutions


X-Ray solutions for electrical components and lighting systems

Additive Manufacturing

Holistic 3D manufacturing inspection for aerospace, medical & automotive

Would you like to learn more about ZEISS Automated Defect Detection?

Contact us. Our experts will get back to you.