analog and digital operations that lead to a terrain classification based on spatial texture in the region of the 
input point. A learning strategy enables the system to refine its processing operations and thus to improve its 
classification accuracy with time. Test results obtained to date are summarized. 
17. Kamphorst, A. and lyer, H.S. 
India 
APPLICATION OF AERIAL PHOTO-INTERPRETATION TO RAVINE 
SURVEYS IN INDIA 
Survey and categorization constitute the first step in the reclamation of lands infested by ravines, which are 
extensive networks of deep gullies formed in alluvial terrain bordering some of the deeply incised rivers of 
North India. Such ravine surveys call for the classification and mapping of ravines under shallow, medium and 
deep classes. 
A morphometric ravine survey adopting a special aerial photointerpretation technique designed to overcome 
the difficulties encountered in measurement of ravine depths and widths is described. The procedure involves 
measurements of depths of ravines at the mid-course of second order ravines from the aerial photographs using 
a parallax bar in conjunction with a mirror stereoscope. The accuracy of these measurements as tested by field 
checks has been found to be plus or minus 1 meter and 1.5 metres respectively on 1:16,000 and 1:30,000 air 
photographs. Widths of ravines are measured on the photos. 
The resulting ravine survey maps indicate economically reclaimable ravines as also the erosion control needs 
of the peripheral lands adjoining the ravines. 
18.  Karpati, I.; Novotny, I. and Varga, G. 
Hungary 
APPLICATION OF THE AERIAL PHOTOGRAPHS IN THE SURVEY OF THE 
BALATON'S PRIMARY PRODUCTION AND IN THE CHANGING OF THE 
MACROPHYTO VEGETATION 
In accordance with the PF section of the International Biological Program, and the aims of the MAB program, 
of the UNESCO we are carrying on research works in order to determine the primary production and the 
variation of the Balaton's macrophyto vegetation. 
In the past, we had made a geodetic survey of the aerial expansion of the reed grass stand, Karpati and Varga, 
1969; Karpati, Varga, and Novotny 1970. This plotting system is labourious and is not accurate. 
A dauer quadrant system 300 x 600 m has been elaborated, i.e. determination of the primary production by 
using aerial photographic interpretation. The complex system includes the procedure of the map making and 
the sampling of the water monolithes as well. 
We are taking photos by hand-held camera from the model area and make use of them in two Ways: 
|) | map making of the vegetation, area and form, by photogrammetric survey; 
2) colour and mottling effects analysis appearing on the photos and determination of the phytocenological 
structures of the associations, isolation of the vegetation, green weight, etc. by interpretation. 
The pattern area is covered by one photography made by a wide-angle camera. 
The favourable photographic height is about 400 m, identical with Lang 1969. During our survey, colour and 
black and white film is applied; Karpati and Varga 1970. We are taking water monolith samples from the 
spots, considered as characteristic ones from the photographs. Based on it, we determine the correlations 
among the dominance, the specific substance and the phyto-biomass production. In this way we can determine 
the total green vegetation. 
The aerial photography and the area samplings, their optimal combination is an expedient method and from 
the application of them, one can minimize the labourious aerial surveys. Such a way, the surveys are more 
79