China
Africa
Explore chapters and articles related to this topic
Recent Trends in OCR Systems: A Review
Published in Amitoj Singh, Vinay Kukreja, Taghi Javdani Gandomani, Machine Learning for Edge Computing, 2023
Aditi Moudgil, Saravjeet Singh, Vinay Gautam
OCR is a process that helps in the identification of characters, either in handwritten or printed text. As shown in the literature, there are many OCR systems that work on the concept of deep learning and use various kinds of feature extraction, segmentation, and classification techniques, along with accuracy measurement techniques, mainly in ancient documents. As reported by many authors, the research in this field is still in its initial stages. The literature presented in this report focuses on various classifications and feature extraction techniques along with the language for which OCR has been designed. The literature focuses on various application areas of the OCR briefly. There is no standard database available for Devanagari manuscripts. Work has already been done on basic characters that can further be extended to modifiers and conjuncts. Lacking behind is the need to design novel OCR for manuscript recognition (Azawi et al., 2013).
Applications of Random Signal Processing
Published in Shaila Dinkar Apte, Random Signal Processing, 2017
Optical character recognition (OCR) is the recognition of printed or handwritten text by a computer. This involves photo scanning of the text character by character, analysis of the scanned image, and then translation of the character image into character codes, such as ASCII. Research in OCR is popular for its various application areas such as office automation, bank check verification, post offices for mail sorting, and a large variety of business and data entry applications. Other applications involve reading aid for the blind, library automation, language processing, and multimedia design.
An Intelligent Whole-Process Medical System Based on Cloud Platform
Published in Applied Artificial Intelligence, 2023
Character recognition technology, also known as optical character recognition (OCR), is a critical component of document digitization. Printed text recognition is the earliest and most mature technology in OCR. German scientist Taushek was granted a patent for OCR in 1929, and since then, European and American countries have been researching Western OCR technology to replace manual keyboard input. After over 40 years of continuous development and improvement, and with the rapid advancement of computer technology, Western OCR technology has become widely used in various fields (Punith, Manish, and Sumanth 2021). This has enabled a vast amount of text materials, such as newspapers, magazines, documents, bills, and reports, to be inputted into computers for efficient processing. The “electronicization” of information processing has thus been realized, resulting in significant savings in time, effort, and costs.
Handwritten optical character recognition by hybrid neural network training algorithm
Published in The Imaging Science Journal, 2019
Nowadays, automatically understanding the text images taken by mobile devices with a digital camera is one of the important vision applications such as understanding the image, automatic sign recognition, and translation, etc. [1]. The cameral-captured text images are easily affected by various problems, such as environmental noise, background complexity, and different deformations. To deal with the above problems, various algorithms are developed for text detection [2–11], recognition [12–23] or end-to-end recognition [24]. Nowadays, many researchers are focusing on the area of text recognition due to various challenges such as lighting, different font size, distortion, blurring [25,26], even the text images are entirely detected. Accordingly, the conventional optical character recognition (OCR) is used for handwritten recognition. Even though the OCR recognition has several problems due to the large intraclass variations, which is caused by unpredictable collections of images and unpredictable background images. Therefore, handwritten OCR is still an important task for many researchers. The classification of handwritten characters is beneficial to fault detection in machinery and medical diagnosis [27,28].
Exploring the usability of the text-based CAPTCHA on tablet computers
Published in Connection Science, 2019
Text-based CAPTCHA is the first and most widely used CAPTCHA. However, it is vulnerable to the bots due to the inclusion of Optical Character Recognition (OCR) elements. Hence, many improvements have been constantly added to this type of CAPTCHA. They include many variations which improve its strength, such as (i) Randomisation of the CAPTCHA length, (ii) Randomisation of the character size, (iii) Waving the CAPTCHA, (iv) anti-recognition elements, (v) anti-segmentation elements. These additional elements make the text-based CAPTCHA less vulnerable to bot attacks. It strengthens the CAPTCHA security, but the solution to the CAPTCHA is more complicated. Hence, it reduces the level of the CAPTCHA usability. From the usability's point of view, the analysis of the impact of these elements which strengthen the security of the text-based CAPTCHA is of great importance for the CAPTCHA users and programmers. Furthermore, the use of the HCI elements, such as the touch screen and its influence to the CAPTCHA usability, is also an interesting point to be explored. All aforementioned is a basis for searching the “ideal” text-based CAPTCHA, which should include the following elements: (i) high usability, (ii) high security, (iii) independency from the users' demographic factors (Pakdel, Ithnin, & Hashemi, 2011).