Shrimp fry counting is an important task for biomass estimation in aquaculture. Accurate counting of the number of shrimp fry in tanks can not only assess the production of mature shrimp but also assess the density of shrimp fry in the tanks, which is very helpful for the subsequent growth status, transportation management, and yield assessment. However, traditional manual counting methods are often inefficient and prone to counting errors; a more efficient and accurate method for shrimp fry counting is urgently needed. In this paper, we first collected and labeled the images of shrimp fry in breeding tanks according to the constructed experimental environment and generated corresponding density maps using the Gaussian kernel function. Then, we proposed a multi-scale attention fusion-based shrimp fry counting network called the SFCNet. Experiments showed that our proposed SFCNet model reached the optimal performance in terms of shrimp fry counting compared to CNN-based baseline counting models, with MAEs and RMSEs of 3.96 and 4.682, respectively. This approach was able to effectively calculate the number of shrimp fry and provided a better solution for accurately calculating the number of shrimp fry.

The rapid development of deep learning has been successfully applied in various fields, including aquaculture, providing new methods and ideas to realize unmanned and intelligent aquaculture. This paper focuses on the technology and research methods used in deep learning for intelligent feeding in fish farming in the past decade, and discusses the application of deep learning in intelligent feeding of fish in detail, including feeding behavior analysis, detection and tracking of live fish, growth state monitoring, residual bait identification and counting, water quality prediction, etc., and summarizes and evaluates the methods, at the same time, analyzes the technical details of the deep learning applied to intelligent precision feeding is analyzed in technical details, including data, algorithms, and evaluation performance indexes. The summarized results show that the advantage of deep learning lies in the automatic extraction of features, which also provides technical support for the construction of intelligent feeding system. However, due to the large differences in fish species, aquaculture environments and data acquisition methods, the technology is less portable, and it is still in the stage of weak artificial intelligence, which requires a large amount of data to train the model, and the cost is high, which has become a bottleneck that restricts the further development of aquaculture. Nevertheless, deep learning has still made breakthroughs in processing complex data. In summary, the purpose of this review is to provide researchers and producers with a better understanding of the current research status of deep learning in intelligent precision feeding of fish, and to provide strong theoretical support for the production process.