Up to now, fishes have shown important roles in our industry, food and pharmacy. Each kind of fish has different significance for human life and nature. The accurate recognition of fishes helps us obtain the benefits of fishes. There exists a large number of species of fishes. Therefore, the accurate recognition of fishes is a challenging issue. The paper presents the detection and recognition of fishes in images using image processing and deep neural networks. We focus on the application of YOLOv8 that is an advanced deep neural network to detect and recognize fishes accurately. The performance evaluation is evaluated on a public datasets of fish images. Obtained results and the performance comparison with existing methods have shown the effectiveness of our proposed method.