Actas de congresos
A Note on Gradient-Based Intensity Normalization
Fecha
2019-01-01Registro en:
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), v. 11867 LNCS, p. 161-169.
1611-3349
0302-9743
10.1007/978-3-030-31332-6_14
2-s2.0-85076082551
Autor
Universidad de Ibagué
Universidade Estadual Paulista (Unesp)
Universidad de Salamanca
Institución
Resumen
This paper presents an improvement on gradient-based intensity normalization introduced by Sintorn et al., which is used in microscopy, especially in the evaluation of proteins and other cell elements in densitometry. In this method, images contain similar textures, but their density and intensity may vary dramatically. Two new profiles are introduced, and the Bhattacharyya distance is employed to find the best matching result. Results show the validity of the improved method.