Fish Species Detection and Tracking Based on Fusion Intensity Entity Transformation using Optical Flow Algorithm

Abstract

Identification of fish species and development aquaculture is support for economic growth in India. By analyzing the commercial development of fish production is large in fishing environment. So monitoring the species in aqua nature and development is important through fish special development progress. The image processing techniques support for species development under the growth movement region. But the problem is quality of analysis is non sophisticated through image analysis results. To resolve this problem, this research study proposes a Fusion Intensity Entity Transformation (FIET) based Optical Flow Algorithm (OFA) to process the images to get accurate result which is recommendation from species growth recommendation. Initially the preprocessing was carried to reduce the noise through Gaussian filters. The segmentation of species is carried out by object enhancement entity identification through enlighten segmentation called Structural Cascaded Object Segmentation (SCOS). Then Fusion Intensity Entity Transformation (FIET) was applied to identify the count species features. Then features get trained with decision flowed optical flow optimization algorithm. This proposed system produce high performance compared to the other system well in detection accuracy.