| In this study a rule based system is designed and developed for the identification, enhancement and extraction of geological lineaments from landsat data. It is divided into three main sections. In the first section, various background topics are covered which include; general information concerning the landsat satellites, elements of image interpretation, general data processing considerations and an overview of the Gulf of Suez.;In the third section, a new rule based algorithm for line and edge detection is developed. Its development makes use of the results obtained from the study of the current algorithms. The characteristic feature of the new algorithm is that it recognizes that for the purpose of line and edge detection a landsat scene needs only to be divided into three classes of data. These classes are lines, slopes, and smooth surfaces. This algorithm is then applied to the landsat data and the resultant images are presented and discussed. It was found that the algorithm successfully removed noise from the scene and was not dependent on the type of data (high frequency, low frequency) within the scene. The algorithm was found to yield results superior to those of the other algorithms currently in use.;In the second section, currently available techniques and methodologies for line and/or edge enhancement are discussed and evaluated. The evaluation is conducted by processing both the landsat data and a developed theoretical data set. The resultant images are then presented and analyzed. The main problems identified were that the quality of the results were dependent on the nature of the data in the scene and the amount of noise in the data. |
