Freshness assessment of tilapia (Oreochromis niloticus) by machine vision based on gill and eye color changes Mega Ayu Yusuf (a*), Setya Permana Sutisna (b)
a Agricultural Engineering Departement, Faculty of Agriculture, Musamus University
b Agricultural Engineering and Biosystem Departement, University of Ibn Khaldun Bogor
Abstract
The freshness of fish is measured using machine vision techniques through changes in the color of the eyes and gills of cultivated tilapia (Oreochromis niloticus), using the levels of brightness (L_), redness (a_), yellowness (b_), chroma (c_), and total color difference parameters (DE) during storage at room temperature. A digital color imaging system, calibrated to provide accurate CIELAB color measurements, was used to record the visual characteristics of the eyes and gills. Regions of interest were selected automatically using a computer program developed in MATLAB software. L_, b_, and DE increased with storage time, while c_decreased. The a_ fisheye parameter does not show a clear trend towards storage time. L_, b_, and DE of fish gills increased with storage time, but a_ and c_ decreased. Regression analysis and an artificial neural network approach were used to correlate eye and gill color parameters with storage time and a strong correlation was found between color parameters and storage time. Changes in gill color are more appropriate than changes in eye color to assess fish freshness. However, the gill covers must be removed to take pictures so this method is destructive and time consuming. Therefore, fish eye color parameters can be used as an environmentally friendly, cheap and easy method to assess fish freshness quickly and online in the food industry.
Keywords: fish colour- fish eye- tilapia- vision machine
