| |
 |
UNIVERSITY OF BUCHAREST
FACULTY OF PHYSICS Guest
2024-11-23 18:31 |
 |
|
|
|
Conference: Bucharest University Faculty of Physics 2023 Meeting
Section: Atomic and Molecular Physics. Astrophysics. Applications
Title:
Artificial intelligence enhanced image analysis in atomic emission spectroscopy
Authors:
Ionuț SLABU, Tudor ȘUTEU, Alexandru NICULESCU, Mihaela COSINSCHI, Vasile BERCU, Leonard Constantin GEBAC
Affiliation:
University of Bucharest, Faculty of Physics, Department of Structure of Matter, Earth and Atmosphere Physics, Astrophysics, 405 Atomistilor Street, Magurele, Ilfov
E-mail
iionutslabu@gmail.com
Keywords:
atomic spectroscopy; Balmer series; artificial intelligence
Abstract:
The observation of the discrete nature of hydrogen's emission spectrum was of foundational importance in developing quantum theory. The atomic model proposed by Niels Bohr explained the empirical formulas developed previously by Balmer. However, in theory, Rydberg's constant, with which one can compute the emission wavelengths, can have two possible values, depending on the approximation used: an infinite mass nucleus system, or a reduced mass system. In this study, we developed an image analysis based protocol that has the sensitivity to discern between the two approximations. The spectrum’s investigation method uses an "in-house" developed software application capable of retrieving the information from a smartphone’s camera and converting it to an array of intensities instead of a matrix made of color channels. Artificial intelligence algorithms (Computer Vision and Deep Convolutional Neural Networks) are used to identify the atom which emits the light in a very efficient, precise and fast way, without any human intervention. The software is packed in a graphical user interface for optimal and easy interaction, cross-platform runtime and suitable for further updates and applications in other fields of physics, e.g., astrophysics.
|
|
|
|