Abstract:
A form of artificial intelligence called machine learning allows a system to learn on its own, without explicit programming or human aid, using its existing knowledge and experience. Based on the given training set, it is used to forecast or make judgments on how to carry out specific tasks. The suggested approach employs a convolutional neural network for picture categorization (CNN). Then, using Tesseract, a recognition engine built on Long Short-Term Memory (LSTM) is used to extract the text from the categorised image. The building blocks of a recurrent neural network are LSTM networks. Through overcoming the overfitting issue, the CNN performs better on very big datasets. Additionally, multiple line text extraction has taken the position of single line text extraction. Thus, adding a big dataset and increasing the number of epochs will increase the accuracy of this system. The number of convolution and pooling layers, as well as the number of nodes in each layer, are chosen using a trial-and-error process. Finally, when compared to other image classification techniques, CNNs employ very less preprocessing.
final year project is based on IEEE Paper.this will be one of the best Final year engineering project for computer science.
Components that we will provide are.
1.complete documentation support
2.complete working hardware/software implemented in students environment
3.classes will held accordingly.