ВИКОРИСТАННЯ ШТУЧНОГО ІНТЕЛЕКТУ ДЛЯ РОЗПІЗНАВАННЯ СКЛАДОВИХ ЕЛЕМЕНТІВ ОБ’ЄКТІВ НА БАЗІ ЗОБРАЖЕННЯ

Abstract

Image recognition is used to retrieve, analyse, understand, and process images from the real world to convert them into digital information. In this area involved data mining, machine learning, pattern recognition, knowledge extension. Developments in the image recognition area have resulted in computers and smartphones becoming capable of mimicking human eyesight. Improved cameras in modern devices can take pictures of very high quality, and with the help of new software, they receive the necessary information and on the basis of the received data is processed images. However, food recognition challenges modern computer vision systems and needs to go beyond just an visible image. Compared to understanding the natural image, visual prediction of ingredients requires high-level solutions and previous knowledge. This creates additional problems, because food components have high variability between the class, when cooking, you have to convert components and the ingredients are often included in the cooked dish. The recognition system allows you to take a step toward understanding the food supply systems such as calorie score and create recipes. The recognition system can be used to address wider problems, such as the prediction of the image on the consistency of the folding elements.