masoud esmailzadeh

Nowadays, different Remote Sensing methods are one of the most practical methods in the field of mineral exploration in mountainous, border areas and without valid geochemical samples and waterway sediments in recognition phase. Spectral methods such as Mixture Tuned Matched Filtering, Spectral Angle Mapper and Linear Spectral Unmixing. The spectrum of each pixel in the image identifies different effects. This study was conducted to determine the areas of alteration, separation of different rocks and minerals, classification of images and finally to determine high potential areas of Ophiolite mineralization, Serpentinites, serpentinized harbors, Listvenites, Chromite and Manganese in Siahcheshmeh and Dizajj areas. In the processing stage, the new Automated Spectral Hourglass method based on spectral methods has been used in order to determine the altered areas and separate the minerals. In order to classify the images, Deep Convolutional Neural Network (CNN), Support Vector Machine (SVM) and Decision Tree J48 (DT) have been used. Finally, the performance of different models was compared based on different evaluation criteria and the CNN model with 98% accuracy compared to the two SVM and DT models with 96% accuracy has been used as the best model for preparing the classification map.

Definition of the efficient ore-forming processes which are considered as mineralization controls is a fundamental stage in mineral prospectivity mapping. In this contribution, five targeting criteria of geochemical, geophysical, geological, structural and hydrothemal alteration data related to epithermal and Carlin-type Au deposits in Takhte-soleyman district, NW Iran, were integrated. For creation of multi-element geochemical layer, principal component analysis was firstly conducted on stream sediment data of 8 selected elements and it was found that PC2 is the representative of Au-As-Sb elemental association in the study area. Then, 10 fuzzified efficient evidence layers were selected based on area under the curve (AUC) of success-rate curves and prepared for generation of predictive models of Au mineralization. For this purpose, two supervised machine learning algorithms, namely multi-layer perceptron (MLP) neural network and support vector machine (SVM) with RBF kernel were used. Comparison of the generated models demonstrates that the latter is more succeeded in delineating exploration targets than the former one.