This document presents research on developing an automated system for fish species detection using deep learning and the MobileNetV2 architecture. The researchers assembled a large dataset of fish photos and used MobileNetV2 for image classification. MobileNetV2 is efficient and effective for this task as it balances accuracy and computational performance. Through training and evaluation, the researchers demonstrated that their method achieves accurate fish species detection and categorization while maintaining computational efficiency compared to other techniques. Their proposed application aims to streamline aquatic ecosystem monitoring and conservation efforts by providing a precise and efficient means of assessing fish populations.