Automated Detection of Tracks in Mountainous Terrain Using IRS-1C/1D Images

Road, railway and track networks are important features of satellite imagery. Automated detection of tracks in mountainous terrain using satellite imagery is a difficult task. The track and the shadow signature of the mountainous terrain are almost same. The presence of ridges and snow also creates a lot of ambiguity. Moreover due to poor sunlight condition in the mountainous terrain the images suffer from poor contrast. So, simple low level operation like, line enhancement processes can not isolate between the track, shadows and ridge signature. The spectral and spatial properties of a satellite image provide information for extracting the line like structures from it. As the low level feature extraction is not sufficient, incorporation of higher level knowledge is necessary. The main contribution of the present track detection method consists of an effective enhancement technique and an efficient segmentation technique that removes from the image, non-track pixels by steps where parameters involved in each step are determined by the sensor characteristic. An efficient linking algorithm based on geometric calculation is proposed. We have tested the algorithm on a number of IRS-1C/1D images (both panchromatic and multispectral) and the results are satisfactory.