Investigationofwhitespotsyndrome inshrimpusingneuralnetworksand deeplearning

Abstract

newline White spot syndrome virus (WSSV) epidemics have seriously harmed penaeid shrimp aquaculture all over the world. There remains an absence of information concerning these complicated viral-host interactions, despite significant attempts to describe the virus, the circumstances that cause infection, and the processes of infection. This understanding is required to develop reliable and efficient treatment strategies for WSSV. The WSSV challenge caused the hepatopancreas to produce more lactic acid, various chemical molecules, as well as other amino acids, while decreasing many amino acids and fatty acids. The rate at which WSSV spread throughout various tissues during infection varied, and the stomach and gill were the most popular locations for WSSV replication in the early stages of infection. The skin, abdomen, and gills are the most heavily affected tissues, which explain why these organs change colour, people consume less, and people gather at the edge of the water because of dyspnea. In this case, image segmentation and classification are used to identify the WSSV shrimp. Mechanisms for segmenting and categorizing images offer a method for extracting features from images based on their objects. Those certain objects are produced using an image segmentation technique in which segments are formed by grouping together pixels with similar spectral properties that are close to one another. The area of interest on any underlying image is protected by image segmentation, a crucial step before actual analysis is recommended in any image processing methodology. In fact, the effectiveness of the segmentation algorithm used will have a big impact on how accurate any image processing performs. This study proposes a typical segmentation technique for segmenting shrimp variability by using essential Canny-GLCM (Gray Level Co-occurrence Matrix) features with a simple Artificial Neural Network (ANN) model.