A Review on Machine Learning and Computer Vision Based Approaches in STEM branch

Ali Ahmadi¹ Iman Bagheri² Somia Molaei³ Seyyed Mostafa Rezaie⁴ Amirhossein Amadeh⁵

1) Department of Civil Engineering, Iran University of Science and Technology, Tehran, Iran
2) Independent Researcher, Mashhad, Iran
3) Department of Computer Engineering, Iran University of Industries & Mines, Tehran, Iran
4) Independent Researcher, Mashhad, Iran
5) Independent Researcher, Mashhad, Iran

Publication : International Congress on Science, engineering & New Technologies(secongress.com/1st)
Abstract :
Machine Learning (ML) and image processing (IP) are both emphasized in this research study, which also discusses the procedures involved in creating digital images and how challenging it is to feed a computer system with them. The first IP stage for this process is acquisition, during which a picture can be loaded and set up for subsequent processing (more on that in the image processing section. One machine learning (ML) approach called neural networks (NN) seeks to improve the answers to the given problem by using input values to predict the output. Well-known companies use NN for maintenance of entire applications, including Facebook for facial detection and recognition, Google for Gmail spam filter, and Microsoft for translation. We have also covered the idea of deep learning, a new development in the field of artificial intelligence (AI), a subfield of machine learning (ML). Since the development of this technology, researchers attention has been drawn to it. We come to a conclusion that although machine learning has effectively penetrated every aspect of computer vision and performs admirably in each one, there are still a few unexplored areas that require further study.
Keywords : image processing computer vision machine learning programming neural networks