Search

Invasive pest fall armyworm (FAW) Spodoptera frugiperda is one of the serious threats to the food safety. Early warning and control plays a key role in FAW management. Nowadays, deep learning technology has been applied to recognize the image of FAW. However, there is a serious lack of training dataset in the current researches, which may mislead the model to learn features unrelated to the key visual characteristics (ring pattern, reniform pattern, etc.) of FAW adults and its close related species. Therefore, this research established a database of 10,177 images belonging to 7 species of noctuid adults, including FAW and 6 FAW close related species. Based on the small-scale dataset, transfer learning was used to build the recognition model of FAW adults by employing three deep learning models (VGG-16, ResNet-50 and DenseNet-121) pretrained on ImageNet. All of the models got more than 98% recognition accuracy on the same testing dataset. Moreover, by using feature visualization techniques, this research visualized the features learned by deep learning models and compared them to the related key visual characteristics recognized by human experts. The results showed that there was a high consistency between the two counterparts, i.e., the average feature recognition rate of ResNet-50 and DenseNet-121 was around 85%, which further demonstrated that it was possible to use the deep learning technology for the real-time monitoring of FAW adults. In addition, this study also found that the learning abilities of key visual characteristics among different models were different even though they have similar recognition accuracy. Herein, we suggest that when evaluating the model capacity, we should not only focus on the recognition rate, the ability of learning individual visual characteristics should be allocated importance for evaluating the model performance. For those important taxonomical traits, if the visualization results indicated that the model didn't learnt them, we should then modify our datasets or adjusting the training strategies to increase the learning ability. In conclusion, this study verified that visualizing the features learnt by the model is a good way to evaluate the learning ability of deep learning models, and to provide a possible way for other researchers in the field who want to understand the features learnt by deep learning models.

For pressurized water reactor nuclear power plants, in order to prevent the release of radioactive substances into environment, fission product barriers (FPBs) are constructed based on the concept of defense-in-depth, including fuel clad, reactor coolant system (RCS), and containment; the status of these FPBs is then acting as an important dimension to decision-making of emergency action levels (EALs). For CPR1000 nuclear power plants, state functions defined in state-oriented emergency operating procedure (SOP) are used to characterize postaccident physical conditions; their degradation substantially represents the challenges on fundamental safety functions and then on the integrity of FPBs in like manner, so degradation of these state functions is referred to as determining initial conditions of each FPB, by which the link between SOP and EALs is established. Then, an intelligent FPB monitoring system (FPBMS) aiming to automatically monitor states of FPBs is developed, verified, and validated. The pioneering work, by building bridges between state functions and initial conditions of FPBs and then computerizing them innovatively, proves that dynamical monitoring of states of FPBs during accident evolvement and real-time indication of loss or potential loss of FPBs can be achieved, which is most helpful in decision-making of EALs.

Interspecific competition is an important factor in the population dynamics and geographical distribution of insect populations. Fall armyworm (FAW), Spodoptera frugiperda, an invasive species, and the Asian corn borer (ACB), Ostrinia furnacalis, a native pest species are major pests on maize in China, posing a threat to maize yield and grain quality. A series of laboratory, greenhouse, and field experiments were conducted to elucidate the competitive interactions between FAW and ACB. In the laboratory experiments, FAW exhibited aggressive behaviors more frequently when compared with ACB, while the latter species exhibited defense behaviors more frequently. Higher intraspecific competition was recorded in the FAW conspecific rather than in ACB, particularly in the 6th larval stage. FAW had a higher interspecific competitive advantage through intraguild predation over ACB, resulting in partial or complete displacement ACB when initially the ratio of the 2 species was 1 : 1. The interspecific competition also had significantly influenced on the population parameters, defensive enzymes, and nutrient of these 2 species. Competitive interaction proved that the response of superoxide, catalase, and soluble protein in FAW were significantly increased, whereas the total sugar content in both species was substantially decreased. Survival rate, and the plant damage that co‐infested by both species varied significantly among the sequential combinations under greenhouse and field conditions. FAW consistently exhibited stronger intraspecific aggression than ACB under laboratory and field conditions when co‐existing on the same feeding guild. These findings contribute to efforts toward the improvement of integrated pest management programs for FAW, in decision making for invasive and native pests' management strategies to reduce the high risks of FAW and ACB outbreaks. [ABSTRACT FROM AUTHOR]