Effective algorithmic operational framework for fish texture evaluation in industry: Achieving maturity

Abstract Reliable, nondestructive fish freshness evaluation applicable during fish commercialization has been continuously pursued by scientists and industry. Taking into account that fish texture is primarily affected even at early stages of post-mortem storage, a relevant nondestructive testing framework for rapid textural assessment of fish freshness was developed in the past. Herein, an algorithm operating within the aforementioned framework and optimized for use in industrial environments is proposed. Sea bass (Dicentrarchus labrax) both freshly killed and stored on ice for 6 days have been used for comparative testing. The fish is part of a system, which is vibration-tested via a new testing protocol designed for easy implementation and robustness to noise. At the same time, a new closed-form analytical expression for the system response to the specific testing is computed and used along with experimental data, for obtaining specific mechanical (thus muscle-structural) characteristics of fish flesh. This computation is designed to only require readily available routines found in most relevant software. The algorithmic operational framework has been used in two different test setups (a custom-built test rig and a prototype device), with results following remarkably similar trends, clearly discriminating different textural (thus freshness) characteristics, and consequently validating the proposed approach.