Robert Desjardins

Map-image matching using a multi-layer perceptron: the case of the road network

Abstract To help automatize map revision at a scale of 1 : 50,000, a map-guided method is described to update the road network of a map database. This paper describes the essential first step of the procedure, which consists of matching the roads present on both the image and the map database. This matching has to be performed precisely in order to generate meaningful hypotheses on the location of new roads. The matching is conducted by using a multi-layer perceptron (MLP) trained to recognize road segments on the SPOT-HRV panchromatic image corresponding to the cartographic database being treated. Two template matching methods using the trained MLP weight matrix are developed. The first method locates all the road intersections on the image, while the second method locates the segments only. Both methods are not accurate enough to be used alone. However, combining both approaches gives results that are reliable enough to be used in the generation of the hypotheses needed to extract new roads.

Atmospheric corrections for remotely-sensed thermal data in a cool humid temperate zone

Abstract This research emphasizes the necessity of carrying out atmospheric corrections in order to obtain accurate absolute temperature values for terrestrial objects, from remotely-sensed data. Satellite data were available for thermal band 6 on Landsat-5. They were complemented by simultaneously-acquired airborne and ground-truth data. Radiosonde data were fed into the computer program LOWTRAN 6. The ground-based data indicated that the use of the LOWTRAN 6 model applied to low-level airborne measurements was very satisfactory (variations of only 0·01-0·61 deg K being noted). However, for the satellite-borne detector the results were much less satisfactory (variations of 0·26-2·05 deg K), because of lower radiometric and spatial resolving power.

SUR LA FORMATION EN TÉLÉDÉTECTION : UN RETOUR AUX SOURCES

SUMMARYDevelopments in remote sensing have for a long time been dominated by a military heritage (pattern recognition) and/or by an attitude of ‘high-tech’ wonderment on the part of the scientific community. Today, efforts are being concentrated on more sophisticated techniques of remote sensing and computerized data processing.While these technical advances are both desirous and fascinating, let us not forget that remote sensing as a perfected tool of observation, analysis, and classification is nothing unless the user has a firm knowledge of the phenomena under investigation and an unequivocably expressed problematic to resolve. The researcher needs to marry intuition, imagination, and memory with concise logic and rigorous method.Any further training or education in remote sensing must be inspired by a symbiosis of empirical and rational knowledge. Furthermore, it will increasingly necessitate the integration of data from several sources, presumably in a ‘user friendly’ environment. Such an environment...