Ms.TEJAL KALE (75) Mr. VIKAS SINGH(72) Mr. HIMANSHU DOYE(30)
PROJECT GUIDE:- Prof. HARSHADA MANDHARE
CONTENTS:- 1. INTRODUCTION 2. AIM 3. OBJECTIVE 4. FEATURES 5. FLOWCHART 6. TECHNOLOGY USED 7. FUTURE SCOPE 8. CONCLUSION AIMs:- Document Image Augmentation is simple to use tool for performing Augmentation Document image analysis and understanding is to automatically recognize and extract textual or graphical material from digitized documents. Document Image Analysis addresses the problem of separation of text and graphics and their identification and recognition. OBJECTIVES:- ● The objective of document image is to recognize the text and graphics components in images, and to extract the intended information as a human would. Two categories of document image analysis can be defined. ● It is a technique that can be used to artificially expand the size of a training dataset by creating modified versions of images. ● It is used to artificially expand the size of a training dataset in order to improve the performance and ability of the model to generalize. FEATURES:- This tool provide following document Augmentation:- 1. Dilate document image 2. Smudge document image 3. Add support for Ground Truth 4. Add Flip and Rotate 5. Add Another Document Specific Augmentation 6. Modify Color Changer to automatically detect background color FLOWCHART:- TECHNOLOGY USED:- PYTHON DEEP LEARNING FUTURE SCOPE / CONCLUSIONS:- Data augmentation has been shown to produce promising ways to increase the accuracy of classification tasks. We experimented with our own way of combining training images allowing a neural net to learn augmentations that best improve the ability to correctly classify images. If given more time, we would like to explore more complex architecture and more varied datasets. To mimic industrial applications, using a VGG16 instead of SmallNet can help us determine if augmentation techniques are still helpful given complex enough networks that already deal with many overfitting and regularization problems. CONCLUSION:- The effectiveness of data augmentation through simple techniques, such as cropping, rotating, and flipping input images. We artificially constrain our access to data to a small subset of the ImageNet dataset, and compare each data augmentation technique in turn.