Recent Trends in Pattern Recognition, Challenges and Opportunities
S. Kannadhasan, R. Nagarajan · 2024
The process of identifying characters that have been subjected to a visible transformation is referred to as optical character identification (OCR). The OCR process allows for the conversion of a wide variety of texts, documents, and digital pictures into an American Standard Code for Information Interchange or another format that is modifiable by a computer. This enables the data to be edited or located. Current developments in pattern recognition have been demanded by a wide variety of applications, such as OCR, document categorization, and data mining, among others. OCR is an essential component of document scanners and plays an important role in the identification of characters and languages, as well as in the protection of financial identities. There are two distinct categories of OCR devices: online character recognition and offline character recognition. Online OCR is superior to offline OCR in terms of accuracy because it handles characters as they are written, bypassing the initial stage of character identification. Offline OCR can be broken down into two categories: printed OCR and handwritten OCR. The process of recording handwritten or typewritten characters into a binary or monochromatic picture, which is then processed by a computer in order to identify the text, is a common method for offline OCR. Scanned documents can now be transformed into text components that computers can recognize, making them more valuable than regular picture files. This was made possible by the introduction of OCR technology. In contrast to the traditional method of manually retyping data into an electronic database, OCR discovers an improved method of automatically inputting data into a database. The most common issue with OCR is the segmentation of associated characters or symbols. The accuracy of the OCR is inversely proportional to the original picture's pixel count.