scholarly journals Research on Deep Learning Techniques in Breaking Text-Based Captchas and Designing Image-Based Captcha

Author(s):  
Janarthanan A ◽  
Pandiyarajan C ◽  
Sabarinathan M ◽  
Sudhan M ◽  
Kala R

Optical character recognition (OCR) is a process of text recognition in images (one word). The input images are taken from the dataset. The collected text images are implemented to pre-processing. In pre-processing, we can implement the image resize process. Image resizing is necessary when you need to increase or decrease the total number of pixels, whereas remapping can occur when you are zooming refers to increase the quantity of pixels, so that when you zoom an image, you will see clear content. After that, we can implement the segmentation process. In segmentation, we can segment the each characters in one word. We can extract the features values from the image that means test feature. In classification process, we have to classify the text from the image. Image classification is performed the images in order to identify which image contains text. A classifier is used to identify the image containing text. The experimental results shows that the accuracy.

2020 ◽  
Vol 2 (2) ◽  
pp. 112-119
Author(s):  
Kawal Arora ◽  
Ankur Singh Bist ◽  
Roshan Prakash ◽  
Saksham Chaurasia

Recent advancements in the area of Optical Character Recognition (OCR) using deep learning techniques made it possible to use for real world applications with good accuracy. In this paper we present a system named as OCRXNet. OCRXNetv1, OCRXNetv2 and OCRXNetv3 are proposed and compared on different identity documents. Image processing methods and various text detectors have been used to identify best fitted process for custom ocr of identity documents. We also introduced the end to end pipeline to implement OCR for various use cases.


2021 ◽  
Vol 3 (2) ◽  
pp. 103-116
Author(s):  
Supriadi Supriadi

The calculator is a calculation tool that is widely used in various specialized fields of business and commerce. The use of a calculator makes it easier for humans to perform arithmetic operations, but there are obstacles in the process of inputting numbers if you want to calculate the value of numbers on written media such as paper, whiteboards and so on. The user must first see the text on written media, then read it and remember it then type the writing on a calculator tool or application. The drawback of this method is that when the user forgets the writing on the written media, the user will see the written text and remember it again so that it takes longer to perform calculations using a calculator. The method used in this study is Optical Character Recognition, this method can recognize text contained in images or handwritten images of mathematical number operations. The results of the text recognition will then be carried out by arithmetic calculations to get the calculation results. From the trials on 20 handwritten images of mathematical number operations, the results obtained were 85% accuracy of extraction and accuracy of handwritten images that can be calculated and correct by 85%


Sign in / Sign up

Export Citation Format

Share Document