Metallic Monuments Image Character Recognition System Using Deep Learning Algorithm

1R.Indra Gandhi , Dr.M.Ponnavaikko

141 Views
38 Downloads
Abstract:

India is known for its wealthy cultural heritage even before prehistoric amount. Recognizing ancient Tamil characters adapt archaeologists to reveal historical events in Cholas amount in twelfth century with vast efforts of the anthropology specialists. The literature were best-known from numerous sources, like inscriptions on temple walls, rocks, pillars, engravings in caves, paintings, copper plates, literature in palm leaves and reports of foreign travelers etc., to write in code their writings. The longer term analysis within the field of archeology can have negative impact thanks to unskillfulness of the manual procedure. Optical Character Recognition practicality is employed to acknowledge ancient Tamil Inscription. OCR module of application i.e. advanced Extreme Deep Learning Machine is especially targeted during this paper. during this paper, we tend to propose a posh extreme Deep learning machine algorithmic rule that this adds some hidden layers to the first ELM network structure, at random initializes the weights between the input layer and therefore the 1st hidden layer also because the bias of the primary hidden layer, utilizes methodology to calculate the parameters of the hidden layers and eventually uses the smallest amount square method to calculate the output weights of the network. The following calculation may be a feature extraction Scale Invariant Feature remodel algorithmic rule to observe and describe native options in input image.

Keywords:

Copper Plate character recognition (CPCR), Region Of Interest (ROI), Complex Extreme Deep Learning Machine (CEDLM), Extreme Learning Machine (ELM), American Standard Code for Information Interchange (ASCII)

Paper Details
Month2
Year2020
Volume24
IssueIssue 8
Pages13415-13420