INNOVATIVE STUDY ON BILSTM-CNN APPROACH FOR FISH FRESHNESS STUDY

Abstract

One of the most crucial aspects to consider when it comes to fish processing, marketing, sales, consumption, and preservation is freshness detection through innovative study. This metric is used to assess how fish have changed as a result of microbiological, chemical, physical, and biochemical factors. Traditional methods for evaluating fish quality are time-consuming and labor-intensive, lacking in-field or real-time applications .The major objective of the paper is to evaluate fish quality and freshness, which is done through image processing. This research recommends a hybrid deep learning model with an automated method based on image processing to assess the freshness of fish. By building a machine learning model using a Bi-directional Long Short Term Memory (Bi

LSTM) and VGG-19 neural network architecture, the approach retrieves features. The fisheye dataset from Kaggle is used as a

sample in this study. The suggested work demonstrates an incremental improvement in fish freshness detection with a 95 percent of accuracy.