An intelligent system for recognition and processing the text with postal addresses
Students Name: Klichuk Serhii Petrovych
Qualification Level: magister
Speciality: Computer Systems and Networks
Institute: Institute of Computer Technologies, Automation and Metrology
Mode of Study: full
Academic Year: 2020-2021 н.р.
Language of Defence: ukrainian
Abstract: Nowadays, there is necessity of text automatic processing instead of manual typing. For instance, one of such solutions is the processing of text from images [1]. This approach solves the following problems: - Save time, because you can significantly reduce the time spent working on this type of employee activity. With this level of optimization, it can be said that employees save more working time to further perform more important tasks. - The quality of the entered data, because if you compare the result of operations entered manually with this technology, such automation makes it almost impossible those mistakes that are made due to the presence of the human factor. - Reduction of budget costs for document processing. This money can be used to improve other processes. - For the client everything works better, faster and error-free, and according to statistics, after such processing, their loyalty increases significantly. Of course, uploading images or documents to a personal computer is often not as convenient as it used to be. In today’s world, all people prefer a mobile gadget - it is always with us wherever we are. Such a pocket assistant can process the text, as well as specified fragments of text from images in the user’s gallery, or instantly take a photo for processing. Qualification work is a developed application that forms a system for analyzing and processing text with email addresses. With this system, you can distinguish between text and identify mailboxes from documents and images. The final product was developed for the Android platform using an integrated development environment (IDE) - Android Studio [2]. The end result of this 8 qualification work is an application based on the android OS [3,4], which implements the following functions: - recognition and analysis of text from the image; - saving text and images that have been recognized by the system in a local database; - image sorting; - confidentiality of information; - recognition of mailboxes in the text; At the stage of technology selection, various architectural solutions were considered and among them the most appropriate one was chosen [5,6]. Based on this, you can easily add new features to the system in the future. With the help of a well-chosen architectural approach, the system processes and stores documents on the user’s side without interaction with the server part, which ensures the confidentiality of all processed data of the user. The input data is a series of black and white images with a size of 1080 * 1080 pixels. The object of research is the research of methods of text recognition from images, use and processing of the received information. Subject of research - in the master’s qualification work methods of text recognition from the image are considered, and also to find the set fragments in the recognized text for acceleration of information processing in bookkeeping, medicine, etc. The purpose of the study: to create a system that speeds up the processing of information from images. Recognize and find previously known fragments of text in documents and images. Reduce economic costs in areas related to mass information processing. In the course of research, all possible analogues of image text processing were developed, and it was found that the alternative proposed by this work is one of the most effective options. 9 List of used literature sources: 1. Image processing methods – [Electronic resource]. – Webpage: https://web.posibnyky.vntu.edu.ua/fksa/2kvetnyj_komp’yuterne_modelyuvannya_s ystem_procesiv/t2/2..htm 2. Official documentation of Android Studio – [Electronic resource]. – Webpage: https://developer.android.com/studio/intro 3. Android operating system – [Electronic resource]. – Webpage: https://uk.wikipedia.org –Access mode: https://uk.wikipedia.org/wiki/Android. 4. Creating applications in Java and Kotlin – [Electronic resource]. – Webpage: http://www.helloworld.ru/texts/comp/lang/java/java5/index.html 5. Robert C. Martin (Uncle Bob) Blog – [Electronic resource]. – Webpage: https://blog.cleancoder.com/uncle-bob/2012/08/13/the-clean-architecture.html 6. Applying Clean Architecture on Android, Hands on – [Electronic resource]. – Webpage: https://five.agency/android-architecture-part-4-applying-clean- architecture-on-android-hands-on/