Open AccessApplication of image analysis technique to determine cleaning of ohmic heating system for milk
Author(s) -
Priyanka Rangi,
P. S. Minz,
Gajanan P. Deshmukh,
Pragaspathy Subramani,
Ripudaman Singh
Publication year - 2019
Publication title -
journal of food science and technology/journal of food science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.656
H-Index - 68
eISSN - 0975-8402
pISSN - 0022-1155
DOI - 10.1007/s13197-019-04011-1
Subject(s) - taguchi methods , joule heating , homogeneity (statistics) , materials science , analytical chemistry (journal) , caustic (mathematics) , ohmic contact , chemistry , mathematics , chromatography , composite material , statistics , layer (electronics) , mathematical physics
Cleaning of equipment is one of the major areas of concern in food industry. Image analysis technique was used to assess the cleaning effectiveness and optimize the CIP protocol for ohmic heating setup. Process parameters selected for optimization of cleaning were caustic concentration (1.0, 1.5, 2.0 and 2.5%), caustic temperature (70, 75, 80 and 85 °C), acid concentration (0.00, 0.25, 0.5 and 0.75%), and acid temperature (40, 50, 60 and 70 °C). Time for caustic treatment was varied from 5 to 20 min at an interval of 5 min, while time acid treatment was kept at a constant of 10 min. Taguchi orthogonal array design was used generate different combinations of acid and alkali concentration and temperature. Images of ohmic heating plates were taken before and after the cleaning procedure. MATLAB program was developed to analyze and extract Gray-Level Co-occurrence (GLCM) matrix properties from the image. Optimized combination was selected based on the highest value of desirability factor among all the experimental set of trials. Treatment with 1.5% caustic concentration at 70 °C for 5 min followed by 0.5% nitric acid concentration at 60 °C was found optimum effective CIP of the heating plates used in ohmic heating setup. GLCM properties correlation, cluster prominence, cluster shade, entropy, homogeneity and inverse difference moment normalized were found suitable for analysis of cleaning effectiveness and optimization of the CIP protocol.