This paper presents the methodology and results of a vegetation reconstruction method based on botanical sampling, the knowledge of succession pattern, digital photograph-interpretation and automatic delineation via image segmentation. The aim is to provide a methodology for interpretation of archived black-and-white aerial photographs, which can be applied at other study sites. Our study area was the Nyíres-tó mire in the Bereg Plain (NE Hungary). Initially, botanical sampling was carried out, and this was followed by separation and identification of current vegetation types. In our study we selected automatic delineation using multi-resolution image segmentation as the method for vegetation mapping. Based on the present-day vegetation map produced and the known successional pathway of the mire, archive aerial photographs were analyzed separately in reverse chronological order to derive plant associations present at the different photograph acquisition dates. With this method we were able to make a chronological sequence of digital vegetation maps over a period of almost fifty years . The analysis of vegetation maps showed that forest cover increased steadily until 1988. After an artificial water supply was introduced (in 1986), the spread of tree-dominated associations became slower, and the relative cover of the different vegetation types reached a stable state.