RESEARCH AND APPLICATION OF CROP PEST MONITORING AND EARLY WARNING TECHNOLOGY IN CHINA

The importance of food security,especially in combating the problem of acute hunger,has been underscored as a key component of sustainable development.Considering the major challenge of rapidly increasing demands for both food security and safety,the management and control of major pests is urged to secure supplies of major agricultural products.However,owing to global climate change,biological invasion(e.g.,fall armyworm),decreasing agricultural biodiversity,and other factors,a wide range of crop pest outbreaks are becoming more frequent and serious,making China,one of the world’s largest country in terms of agricultural production,one of the primary victims of crop yield loss and the largest pesticide consumer in the world.Nevertheless,the use of science and technology in monitoring and early warning of major crop pests provides better pest management and acts as a fundamental part of an integrated plant protection strategy to achieve the goal of sustainable development of agriculture.This review summarizes the most fundamental information on pest monitoring and early warning in China by documenting the developmental history of research and application,Chinese laws and regulations related to plant protection,and the National Monitoring and Early Warning System,with the purpose of presenting the Chinese model as an example of how to promote regional management of crop pests,especially of cross border pests such as fall armyworm and locust,by international cooperation across pest-related countries.

Monitoring Thosea sinensis Walker in Tea Plantations Based on UAV Multi- Spectral Image

Thosea sinensis Walker(TSW)rapidly spreads and severely damages the tea plants.Therefore,finding a reliable operational method for identifying the TSW-damaged areas via remote sensing has been a focus of a research community.Such methods also enable us to calculate the precise application of pesticides and prevent the subsequent spread of the pests.In this work,based on the unmanned aerial vehicle(UAV)platform,five band images of multispectral red-edge camera were obtained and used for monitoring the TSW in tea plantations.By combining the minimum redundancy maximum relevance(mRMR)with the selected spectral features,a comprehensive spectral selection strategy was proposed.Then,based on the selected spectral features,three classic machine learning algorithms,including random forest(RF),support vector machine(SVM),and k-nearest neighbors(KNN)were used to construct the pest monitoring model and were evaluated and compared.The results showed that the strategy proposed in this work obtained ideal monitoring accuracy by only using the combination of a few optimized features(2 or 4).In order to differentiate the healthy and TSW-damaged areas(2-class model),the monitoring accuracies of all the three models were computed,which were above 96%.The RF model used the least number of features,including only SAVI and Bandred.In order to further discriminate the pest incidence levels(3-class model),the monitoring accuracies of all the three models were computed,which were above 80%,among which the RF algorithm based on SAVI,Band_(red),VARI__(green),and Band_(red_edge) features achieve the highest accuracy(OAA of 87%,and Kappa of 0.79).Considering the computational cost and model accuracy,this work recommends the RF model based on a few optimal feature combinations to monitor and distinguish the severity of TSW in tea plantations.According to the UAV remote sensing mapping results,the TSW infestation exhibited an aggregated distribution pattern.The spatial information of occurrence and severity can offer effective guidance for precise control of the pest.In addition,the relevant methods provide a reference for monitoring other leaf-eating pests,effectively improving the management level of plant protection in tea plantations,and guaranting the yield and quality of tea plantations.

Empowering fall webworm surveillance with mobile phone-based community monitoring: a case study in northern China

Recent advances in information and communication technologies, such as mobile Internet and smart- phones, have created new paradigms for participatory environment monitoring. The ubiquitous mobile phones with capabilities such as a global positioning system, camera, and network access, offer opportunities to estab- lish distributed monitoring networks that can perform a wide range of measurements for a landscape. This study examined the potential of mobile phone-based community monitoring of fall webworm （Hyphantria cunea Drury）. We built a prototype of a participatory fall webworm monitoring System based on mobile devices that stream- lined data collection, transmission, and visualization. We also assessed the accuracy and reliability of the data collected by the local community. The system performance was evaluated at the Ziya commune of Tianjin municipality in northern China, where fall webworm infestation has occurred. The local community provided data with accuracy comparable to expert measurements （Willmott＇s index of agreement 〉0.85）. Measurements by the local community effectively complemented remote sensing images in both temporal and spatial resolution.

Precise Agriculture:Effective Deep Learning Strategies to Detect Pest Insects

Pest insect monitoring and control is crucial to ensure a safe and profitable crop growth in all plantation types,as well as guarantee food quality and limited use of pesticides.We aim at extending traditional monitoring by means of traps,by involving the general public in reporting the presence of insects by using smartphones.This includes the largely unexplored problem of detecting insects in images that are taken in noncontrolled conditions.Furthermore,pest insects are,in many cases,extremely similar to other species that are harmless.Therefore,computer vision algorithms must not be fooled by these similar insects,not to raise unmotivated alarms.In this work,we study the capabilities of state-of-the-art(SoA)object detection models based on convolutional neural networks(CNN)for the task of detecting beetle-like pest insects on nonhomogeneous images taken outdoors by different sources.Moreover,we focus on disambiguating a pest insect from similar harmless species.We consider not only detection performance of different models,but also required computational resources.This study aims at providing a baseline model for this kind of tasks.Our results show the suitability of current SoA models for this application,highlighting how FasterRCNN with a MobileNetV3 backbone is a particularly good starting point for accuracy and inference execution latency.This combination provided a mean average precision score of 92.66%that can be considered qualitatively at least as good as the score obtained by other authors that adopted more specific models.