direct counterattack _Crossen_
or why I no longer post photos of political graffiti:
Yes, security authorities and intelligence services have highly sophisticated tools to identify handwriting and graffiti in 2026. In cases of politically motivated acts (such as the right-wing code 1161 described above), the state security police usually take over the investigations, while intelligence services (such as the Federal Office for the Protection of the Constitution) analyze the underlying structures. The evaluation is based on a combination of human expertise, central databases and modern AI:
1. The Database: The "FISH" System
The Federal Criminal Police Office (BKA) utilizes the world-leading **Forensic Information System for Handwriting (FISH)**.
- **How it works:** A photograph of the wall inscription is scanned. Interfering backgrounds (such as the texture of the plaster) are digitally filtered out until only the pure lettering remains.
- **The comparison:** The system extracts geometric features completely independent of language or content. It then compares these with thousands of samples from known offenders, other crime scenes, or seized materials (e.g., notebooks from house searches).
2. The Machines: AI and Deep Learning in 2026
In 2026, **Artificial Intelligence (AI)** is deeply integrated into forensic image analysis. Deep-learning systems (Convolutional Neural Networks) analyze characteristics that are barely measurable by the human eye:
- **Pressure and stroke behavior:** How fast was the paint sprayed? Where is the layer of paint thicker (because the sprayer paused briefly)?
- **Micro-features:** The slant angle of the lines, the spacing between the digits, and the exact curvature of the "1" or "6". The AI can calculate whether the red "1" and the white "1116", despite the different colors, could anatomically originate from the same hand.
3. The Human Factor: Forensic Handwriting Experts
Despite all AI support, the final word always rests with **human handwriting experts**. Machines provide probabilities and filter out suspects, but a certified expert must prepare the official report for court proceedings.
Sprayers often have an unconscious, deeply ingrained "writing flow" (their motor habit) that they can hardly conceal, even when deliberately trying to disguise their handwriting.
The Biggest Problem for the Perpetrators: The "Tagger Profile"
When dealing with graffiti, authorities rarely hunt for a single isolated image. Police forces maintain extensive **graffiti databases** (such as the _INGRID_ system used for research or police-specific investigative databases).
Since sprayers are usually repeat offenders who leave the same code or "tag" dozens of times across a city, investigators link these locations together. Combined with cell tower data dumps (which smartphones were logged into the network near the wall at the time of the offense?) or nearby video surveillance, the "wall battle" on the street often leads directly to the identity of the perpetrators.
direct counterattack _Crossen_
or why I no longer post photos of political graffiti:
Yes, security authorities and intelligence services have highly sophisticated tools to identify handwriting and graffiti in 2026. In cases of politically motivated acts (such as the right-wing code 1161 described above), the state security police usually take over the investigations, while intelligence services (such as the Federal Office for the Protection of the Constitution) analyze the underlying structures. The evaluation is based on a combination of human expertise, central databases and modern AI:
1. The Database: The "FISH" System
The Federal Criminal Police Office (BKA) utilizes the world-leading **Forensic Information System for Handwriting (FISH)**.
- **How it works:** A photograph of the wall inscription is scanned. Interfering backgrounds (such as the texture of the plaster) are digitally filtered out until only the pure lettering remains.
- **The comparison:** The system extracts geometric features completely independent of language or content. It then compares these with thousands of samples from known offenders, other crime scenes, or seized materials (e.g., notebooks from house searches).
2. The Machines: AI and Deep Learning in 2026
In 2026, **Artificial Intelligence (AI)** is deeply integrated into forensic image analysis. Deep-learning systems (Convolutional Neural Networks) analyze characteristics that are barely measurable by the human eye:
- **Pressure and stroke behavior:** How fast was the paint sprayed? Where is the layer of paint thicker (because the sprayer paused briefly)?
- **Micro-features:** The slant angle of the lines, the spacing between the digits, and the exact curvature of the "1" or "6". The AI can calculate whether the red "1" and the white "1116", despite the different colors, could anatomically originate from the same hand.
3. The Human Factor: Forensic Handwriting Experts
Despite all AI support, the final word always rests with **human handwriting experts**. Machines provide probabilities and filter out suspects, but a certified expert must prepare the official report for court proceedings.
Sprayers often have an unconscious, deeply ingrained "writing flow" (their motor habit) that they can hardly conceal, even when deliberately trying to disguise their handwriting.
The Biggest Problem for the Perpetrators: The "Tagger Profile"
When dealing with graffiti, authorities rarely hunt for a single isolated image. Police forces maintain extensive **graffiti databases** (such as the _INGRID_ system used for research or police-specific investigative databases).
Since sprayers are usually repeat offenders who leave the same code or "tag" dozens of times across a city, investigators link these locations together. Combined with cell tower data dumps (which smartphones were logged into the network near the wall at the time of the offense?) or nearby video surveillance, the "wall battle" on the street often leads directly to the identity of the perpetrators.