Multi-Digit Number Classification using MNIST and ANN

Modern machines have difficulty in reading the handwritten numbers, as every person's handwriting is different and unique. However, in the modern era where everyone is shifting towards digital technology. Retyping of handwritten ledgers and documents into computer systems will be a hefty process. Thanks to Artificial Intelligence computer vision techniques, that has eased the path. Thus, we have developed an Artificial Neural Network System that can read any type of handwritten number with an accuracy percentage of more than 93. Our Project consists of both research work and practical implementation. The Neural Network tests and learns with the help of MNSIT Data after the successful completion of the code we get a system that can read and process the images with an accuracy of 93% which is also supported by the graph. We have stored this system in a file and now we can use this system and with the help of some libraries like opencv, etc. We can read and save the processed output in a CSV file in just few clicks and get the data in real-time. The project is in its earlier stages. Once finalized it can help in saving a lot of Human Efforts. Keywords— Artificial Intelligence, Artificial Neural Networks, Handwritten Number, Machine Learning, Deep Learning, MNIST

[1]  Raveendran Paramesran,et al.  An efficient method for the computation of Legendre moments , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Toshio Ito,et al.  Simulation of Detecting Function object for AGV Using Computer Vision with Neural Network , 2016, KES.

[3]  Ompriya Kale,et al.  A Survey on Feature Extraction Methods for Handwritten Digits Recognition , 2014 .

[4]  Cicero Ferreira Fernandes Costa Filho,et al.  Handwritten Digit Recognition Using SVM Binary Classifiers and Unbalanced Decision Trees , 2014, ICIAR.

[5]  Saeed Al-Mansoori,et al.  Intelligent Handwritten Digit Recognition using Artificial Neural Network , 2015 .

[6]  Mounir Ait Kerroum,et al.  Feature Extraction based on DCT for Handwritten Digit Recognition , 2014 .

[7]  Andreas Geiger,et al.  Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art , 2017, Found. Trends Comput. Graph. Vis..

[8]  Venu Govindaraju,et al.  Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks , 2016, ICML.

[9]  Ezzat El-Sherif,et al.  Arabic handwritten digit recognition , 2008, International Journal of Document Analysis and Recognition (IJDAR).

[10]  Alireza Khotanzad,et al.  Invariant Image Recognition by Zernike Moments , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  L. Deng,et al.  The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web] , 2012, IEEE Signal Processing Magazine.

[12]  Ming-Kuei Hu,et al.  Visual pattern recognition by moment invariants , 1962, IRE Trans. Inf. Theory.

[13]  Miroslaw Pawlak,et al.  On Image Analysis by Moments , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Mahdi Jampour,et al.  Efficient Handwritten Digit Recognition based on Histogram of Oriented Gradients and SVM , 2014 .

[15]  Alexander Kadyrov,et al.  Affine invariant features from the trace transform , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Alex M. Andrew,et al.  Object Recognition in Man, Monkey, and Machine , 2000 .

[17]  Ram Gopal Raj,et al.  Recognition of Traffic Sign Based on Bag-of-Words and Artificial Neural Network , 2017, Symmetry.

[18]  Parshuram M. Kamble,et al.  Handwritten Marathi character recognition using R -HOG Feature , 2015 .

[19]  Abdelhak Boukharouba,et al.  Novel feature extraction technique for the recognition of handwritten digits , 2017 .