A novel framework based on deep neural network fordetermining the melting point of crystalline chemical substances
Shrivastava, Anurag (Bennett University (Índia))
Shrivastava, Bhavana (DIT University (Índia))

Date:	2024
Abstract:	The melting point detection apparatus is a tool employed in the pharmaceutical and chemical industriesto ascertain the melting point of chemical substances. It plays a crucial role in assessing the quality andclassification of these substances by identifying their purity or impurities. The proposed framework, whichis built upon a deep learning model, utilizes a deep neural network (DNN) incorporating tensorlow, keras,and activation functions. This framework's primary purpose is to classify images representing the stateof crystalline chemicals and determine their respective melting points. The class and sub-class labels ofcrystalline chemical's states were taken into consideration by this deep learning model as a knowledge,which can limit the distance between features in distinct crystalline chemical states images. Additionally, arobust activation function that can fully reserve the image edge features of the chemical's melting region wasemployed to fit tolerance complete slide image recognition. In the proposed framework activation functionswere applied with DNN for the most accurate image classification training purpose. The experiment findingsdemonstrate that the suggested technique has strong resilience and generalisation, which results in a greaterclassification accuracy (up to 99. 7%) and offers the proposed framework as an effective tool for meltingpoint determination.
Rights:	Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, i la comunicació pública de l'obra, sempre que no sigui amb finalitats comercials, i sempre que es reconegui l'autoria de l'obra original. No es permet la creació d'obres derivades.
Language:	Anglès
Document:	Article ; recerca ; Versió publicada
Subject:	Deep Neural Network ; Tensorflow ; Deep Learning ; Convolution Neural Network ; Keras ; Rasp-berry pi
Published in:	ELCVIA. Electronic letters on computer vision and image analysis, Vol. 23 Núm. 1 (2024) , p. 47-67 (Regular Issue) , ISSN 1577-5097

Adreça original: https://elcvia.cvc.uab.cat/article/view/1527
Adreça alternativa: https://raco.cat/index.php/ELCVIA/article/view/980000001026
DOI: 10.5565/rev/elcvia.1527

18 p, 571.9 KB

The record appears in these collections:
Articles > Published articles > ELCVIA
Articles > Research articles