Scientists have found a way to spot fake Van Goghs

The method is described in the journal *Surface Topography: Metrology and Properties*.

The essence of the method is to analyse high-quality images of a painting and convert them into a surface map: where the brushstrokes are higher, where they are lower, and how complex and uneven the paint’s texture is.

The researchers tested the approach on works associated with Vincent van Gogh. The well-known forgery The Plowmen differed sharply from the authenticated paintings, whilst the work Sunset at Montmajour, previously recognised as genuine, turned out to be closer to the artist’s real works.

Details

Art forgeries often attempt to replicate the subject matter, colours and general style of the artist. But it is much harder to fake the minute physical characteristics of the brushstrokes: how the paint sits on the canvas, how uneven the surface turns out, and how the brush movements are repeated.

This is precisely what the scientists set out to measure. They used high-resolution images of the paintings’ surfaces and analysed them as complex reliefs. In essence, the painting was viewed not only as an image, but also as a three-dimensional surface.

Fractal analysis was used to assess this surface. Put simply, this is a way of measuring how complex and ‘uneven’ a pattern looks at different scales. If an artist paints in a characteristic style, their brushstrokes may leave a repeating relief pattern — a sort of morphological signature.

In tests, the method was able to distinguish works attributed to Van Gogh from a well-known forgery. *The Ploughmen* stood out from the rest, whilst *Sunset at Montmajour* showed greater similarity to the artist’s authenticated works.

The authors also compared Van Gogh’s characteristics with the works of the 17th-century artist David Kløcker Erensthal. The method was able to distinguish between different artistic ‘handwritings’, demonstrating its potential usefulness beyond Van Gogh’s legacy.

However, the technology does not resolve all issues on its own. The authenticity of a painting is usually verified through a comprehensive process: examining the history of ownership, archival documents, the canvas, pigments, ground, signature, the condition of the paint layers and traces of restoration. The new method adds another level of verification — the analysis of the surface of the brushstrokes.

Why this matters

For museums and collectors, forgery is not just a financial risk. It is also a risk to cultural memory: a fake work may end up in an exhibition, a catalogue or a scholarly work and distort the perception of the artist.

The method is interesting in that it does not require taking a paint sample or physically interfering with the painting. This is particularly important for valuable works: any damage, even minimal, may be unacceptable.

If the technology proves reliable on large samples, it will be able to help experts spot suspicious works more quickly. This is particularly important in the art market, where works by famous artists are worth millions and the pressure on experts and buyers is very high.

But the main advantage is not that the computer will ‘replace’ humans. It is something else entirely: digital analysis can provide measurable data where previously an expert saw only visual similarities or doubts.

Background

Attributing paintings is one of the most complex tasks in art history. Even if a work resembles the style of a famous artist, that is not enough. It is necessary to understand when and by whom it was created, what materials were used, whether the technique corresponds to the period, and whether there is a documented history of the work.

In recent years, digital methods have increasingly been added to traditional expertise: image analysis, machine learning, the study of brushstroke structure, X-rays, infrared imaging and chemical analysis of materials. Algorithms for recognising Van Gogh’s works based on brushstroke characteristics and digital image features have been proposed previously; for example, a study on arXiv described a method for classifying genuine and forged Van Gogh paintings using statistical features of directional textures.

The new work differs in that it focuses on topography — that is, the surface relief. This is particularly important in painting, where a brushstroke is not only colour but also a physical trace of the brush’s movement.

Source

Study: Preserving Van Gogh’s Painterly Heritage: Topographical and Fractal Insights in Authentication, journal Surface Topography: Metrology and Properties, 2026.