Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/2781
Full metadata record
DC FieldValueLanguage
dc.contributor.authorGorai, Amit Kumar-
dc.contributor.authorPatel, Ashok Kumar-
dc.contributor.authorChatterjee, Snehamoy-
dc.date.accessioned2017-11-14T11:14:04Z-
dc.date.available2017-11-14T11:14:04Z-
dc.date.issued2017-11-
dc.identifier.citation7th Asian Mining Congress, Kolkata, India, 8-11 November 2017en_US
dc.identifier.urihttp://hdl.handle.net/2080/2781-
dc.descriptionCopyright of this paper belongs to proceedings publisheren_US
dc.description.abstractThe quality control in metal mines is always a challenging task due to complex nature of ore reserves. The present study attempts to develop adaptive neuro-fuzzy inference system (ANFIS) for classification of iron ores. The ANFIS system was developed using the optimised image feature set as input and the ore classes as the output. The study used 812 image samples for feature extractions. The sample data were partitioned for training and testing in the ratio of 70:30 respectively. The performance of the ANFIS system was evaluated using four confusion matrix parameters viz., sensitivity, specificity, misclassification, and accuracy, which were found to be 0.8750, 0.9681, 0.0510, and 0.9490 respectively. The high value of sensitivity, specificity, and accuracy, and the low value of the misclassification indicate a good performance of the model. It was observed that 13 % of the total testing image samples was misclassified by the model. Thus, the proposed model can be used satisfactorily for classification of iron ores.en_US
dc.subjectANFISen_US
dc.subjectIron Oreen_US
dc.subjectVision Based Systemen_US
dc.titleAdaptive Neuro-Fuzzy Inference System (ANFIS) For Classification of Iron Ore Using Vision Based Systemen_US
dc.typeArticleen_US
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2017_7thAMC_AKGorai_Adaptive.pdf282.99 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.