RETRACTED ARTICLE: An optimized deep network-based fish tracking and classification from underwater images

Underwater imagery detection, classification and analysis is one of the essential parts of marine technology as well as fisheries management. Also, fishery science management has utilized stock assessment and statistical algorithms that are inefficient and prone to human errors. Various computer based fish tracking algorithms are provides solution but that are not optimal solution so many drawbacks still exist. This current research article proposes a novel chimp-based Google Deep Network (CbGDNet) to detect and classify the fish from the underwater image dataset. Initially, an aquatic image database is collected through the net source and passed into the proposed system for further processing. Moreover, unwanted data is removed in the pre-processing module and filtered data is collected. After that, feature extraction phase is enabled to extract the necessary features from the filtered data. Moreover, the proposed model effectively tracks and classifies fish and types from the image database. The proposed model is validated using a sea animal image dataset, and the results are evaluated. The performance of the proposed model shows that the proposed method obtained 99.16% accuracy. Additionally, a comparative analysis is conducted to verify the effectiveness of the designed model. Hence, it proved that the presented technique gained better outcomes.