Target detection technology plays a crucial role in fishery ecological monitoring, fishery diversity research, and intelligent aquaculture. Deep learning, with its distinct advantages, provides significant convenience to the fishery industry. However, it still faces various challenges in practical applications, such as significant differences in image species and image blurring. To address these issues, this study proposes a multi-scale, multi-level, and multi-stage cross-domain feature fusion model. In order to train the model more effectively, a new dataset called Fish52 (Multi-scene Fish dataset) was constructed, on which the model achieved an mAP of 82.57%. Furthermore, we compared prevalent one-stage and two-stage detection methods on the Lahatan (single-scene fish dataset) and Fish30 dataset and tested them on the F4k and FishNet dataset. The mAP of our proposed model on the Fish30, Lahatan, F4k, and FishNet datasets reaches 91.72%, 98.7%, 88.6%,and 81.5% respectively, outperforming existing mainstream models. Comprehensive empirical analysis indicates that our model possesses high generalization ability and has reached advanced performance levels. In this study, the depth of the model backbone is deepened, a novel neck structure is proposed, and anew module is embedded therein. To enhance the fusion ability of the model, a new attention mechanism module is introduced. In addition, in the adaptive decoupling detection header module, introducing classes with independent parameter sand regression adapters reduces interaction between different tasks. The proposed model can better monitor fishery resources and enhance aquaculture efficiency. It not only provides an effective approach for fish detection but also has certain reference significance for the identification of similar target sin other environments, and offers assistance for the construction of smart fisheries and digital fisheries.

The low and unstable performance is still one of the major obstacles for anaerobic digestion, which provokes researchers to develop some strategies to improve it. In addition to traditional approaches such as co-digestion, pretreatment, and recirculation, some emerging strategies, namely additive and microaeration have also been recognized and developed in recent years due to the update of knowledge. Many studies have evaluated the effect of those strategies on digestion performance. However, it lacks a unified analysis and induction, especially the mechanism for different strategies. Therefore, this review presents a comprehensive view of research progress on the strategies based on latest researches considering five main strategies. Critical think and summary of their mechanism, reactor performance, and availability of those strategies were executed. The results demonstrated that the contribution of microaeration is mainly to balance the composition and activity of hydrolysis, acidogenesis, and methanogenic archaea. Recirculation and co-digestion balance mass and reaction environments. Pretreatment makes the structure of the substrate (such as removing lignin, reducing cellulose crystallinity, and increasing the substrate-specific surface area) more conducive to microbial utilization. The mechanism of the additive strategy varies greatly depending on the type of additive, such as enhancing interspecies electrons transfer through conductive materials, resisting adverse digestion conditions through functional microbial additives, and accelerating nutrient absorption by regulating the bioavailability of trace elements. Although these strategies have different mechanisms for promoting digestion performance, their ultimate goal is to make the parameters of the reactor to an ideal status, and then to achieve a balance among substance, microorganisms and water in anaerobic reactor.