Copper Plate Image Character Recognition System Using Complex Extreme Deep Learning Machine

Abstract

Recognizing ancient Tamil characters enable archaeologists to reveal historical events in Cholas period in 12th century with huge efforts of the archaeological experts. The future research in the field of archaeology will have negative impact due to inefficiency of the manual procedure. Optical Character Recognition (OCR) functionality is used to recognize ancient Tamil Inscription. OCR module of application i.e. Complex Extreme Deep Learning Machine is mainly focused in this paper. In this paper, we propose a Complex Extreme Deep learning machine calculation (CEDLM) that the CEDLM adds some concealed layers to the first ELM organize structure, arbitrarily introduces the loads between the information layer and the primary shrouded layer just as the predisposition of the principal concealed layer, uses the technique (make the real each concealed layer yield approach the normal each concealed layer yield) to ascertain the parameters of the covered up layers(except the main concealed layer), lastly utilizes the least square strategy to figure the yield loads of the system. The consequent count is an element extraction (Scale Invariant Feature Transform (SIFT) calculation to identify and portray nearby highlights in pictures. Correlation with the trial consequences of different procedures uncovered the capability of the proposed framework and showed that the component choice methodology expanded the precision of the grouping procedure.