• P. Darwin
• Pedagandham N Sai Siri Srujana
• Danduprolu Devi SriHarika
• Mohammad Raheem Sharief

In aquatic ecosystems and fishery management, precise species identification is all that counts. However, the existing Support Vector Machine (SVM) based models have several limitations such as slow processing speed and low accuracy in dealing with complex data. Additionally existing fish classifiers using Alex Net, tend to have an accuracy of 95.5%. In order to overcome these challenges and limitations, we propose a new fish classification system built on deep learning and the CNN (Convolutional Neural Network) model resulting in a significant boost in accuracy. As a result, this system self-explains complex themes and structures from fish imagery thus increasing classification precision. Also, our system measures accuracy aspects including precision, recall and F1 score which gives a high reliability of results. It allows to improve the fish species identification, avoid misclassifications and prove data accuracy. The Automated Fish Species Identification system employs advanced technologies such as deep learning and CNN to analyse data mainly presented through images. The species of fish can be detected quickly and accurately, as a result of this many activities including those of monitoring fish populations, evaluation of ecosystem health assessment, sustainable fisheries management and conservation of aquaculture biodiversity etc. can depend upon having these technologies embedded in their systems. By using the automatic identification process, our system is fast and economical in terms of saving time and resources while providing accurate data that can be used for sound decision-making concerning aquatic environment.

CNN, Key metrics, SVM model, Marine Research.

"An Automated Fish Classification System using CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 2, page no.e443-e448, February-2024, Available :http://www.jetir.org/papers/JETIR2402465.pdf

"An Automated Fish Classification System using CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 2, page no. ppe443-e448, February-2024, Available at : http://www.jetir.org/papers/JETIR2402465.pdf