Abstract

Digital Image Processing is an expeditiously emerging field possessing a large number of applications in science and engineering aspects. One of the most used applications in almost every sector is Optical Character Recognition (OCR). OCR is the electronic conversion of handwritten text into digital format which makes information processing from printed papers to data records easy, thus helping to electronically edit, search and store printed texts into machines. This text can then be used in variety of applications like machine translation, speech-to-text, pattern recognition etc. OCR as a piece of software applies pre-processing to improve the recognition in images. This pre-processing step includes skewness correction, despeckling, layout analysis and line and word detection. OCR saves tons of manual effort by recognizing handwritten text with word level detection resulting in an accuracy of 81% to 90%. With form processing, one can capture information in digital format that can save time, labor and money. This helps in achieving a better accuracy in detection. Such systems range from minor application forms to large scale survey forms. Deep Learning algorithms dealing with computer vision related tasks can be used to build a recognition engine.

References

• Shailendra Singh Kathait and Shubhrita Tiwari. Application of Image Processing and Convolution Networks in Intelligent Character Recognition for Digitized Forms Processing. Valiance Solutions Pvt. Ltd., Noida, Uttar Pradesh, 2018
• Dipti Deodhare, NNR Ranga Suri and R. Amit. Preprocessing and Image Enhancement Algorithms for a Form-based Intelligent Character Recognition System. Technomathematics Research Foundation, 2005
• Maureen Caudill. Neural Network Primer: Part I. AI Expert, Feb. 1989
• Ian Goodfellow and Yoshua Bengio and Aaron Courville. Deep Learning MIT Press, http://www.deeplearningbook.org, 2016
• Alex Krizhevsky, Geoffrey E. Hinton and Ilya Sutskever. ImageNet Classification with Deep Convolutional Neural Networks. University of Toronto.
• Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, Zbigniew Wojna Rethinking the Inception Architecture for Computer Vision https://arxiv.org/abs/1512.00567
• Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich Going Deeper with Convolutions
• Karen Simonyan, Andrew Zisserman Very Deep Convolutional Networks for Large-Scale Image Recognition https://arxiv.org/abs/1409.1556
• ImageNet: 2016 Stanford Vision Lab, Stanford University, Princeton University http://www.image-net.org/
• Tensorflow https://www.tensorflow.org/
• David G. Lowe Distinctive Image Features from Scale-Invariant Keypoints https://www.cs.ubc.ca/~lowe/papers/ijcv04.pdf
• Jie Hu, Li Shen, Samuel Albanie, Gang Sun, Enhua Wu Squeeze-and Excitation Networks https://arxiv.org/abs/1709.01507
• Lisa Torrey and Jude Shavlik Transfer Learning, Handbook of Research on Machine Learning Applications, 2009. http://ftp.cs.wisc.edu/machine-learning/shavlik-group/torrey.handbook09.pdf
• Django https://www.djangoproject.com/
• Amazon Web Services https://aws.amazon.com/ec2/
• Artifical Neural Network https://en.wikipedia.org/wiki/Artificial neural network.
• Fjoder Van Veen The Neural Network Zoo https://www.asimovinstitute.org/neural-network-zoo/
• Google Cloud https://cloud.google.com/tpu/docs/inception-v3-advanced