FMC and SenseAgro jointly release in China an applet for identifying the fall armyworm

Date:07-06-2020

The fall armyworm (FAW), or Spodoptera frugiperda, is a migratory pest warned by FAO globally. Since January, 2019 when it began to invade Yunnan, China, the pest has been founded on 18 crops in 27 provinces including Taiwan in half a year. According to the statistics of the National Agro-Tech Extension and Service Center, since May, 2020, the pest has caused damage to 3.87 million mu of farmland, and the spread is accelerating, particulary to the northern part of the country, resuling in increasing area of impacted crops. The prevention and control situation is simply tough.

Recently, FMC, the world's leading agricultural technology company, and Shenzhen SenseAgro Technology Co., Ltd., an agricultural artificial intelligence (AI) technology developer, jointly developed and launched a "Field FAW Identification System" using AI technology, taking the lead in exploring the combination of AI and FAW control. On July 3, 2020, the system was launched at an online press conference held by FMC and SenseAgro. The conference also invited experts from Institute of Agricultural Genomics, Chinese Academy of Agricultural Sciences (Shenzhen), and  Institute of Plant Protection, Chinese Academy of Agricultural Sciences, to explain the evolution of fall armyworm and the control methods, attracting more than 40,000 people online watching. 

Tracy Wu，President and General Manager of FMC China, Business Director of FMC North Asia, addressed in the conference

Lu Weixin (left), strategic project manager of FMC Great China, and Xie Qiufa, CEO of SenseAgro, signed a cooperation agreement at the press conference

AI-based precision identification helps meet the needs of FAW control

According to China Statistical Yearbook 2019, there are 721.35 million rural people in China, of whom about 260 million earn a living on farming, and nearly 98% of them are scattered small growers. At present, the FAW in China mainly feeds on corn, which is highly disruptive and reproductive and extremely harmful to corn crop. The contradiction between the pressing demand for FAW control and the shortage of crop protection experts has become the biggest pain point for farmers to secure a good harvest.

To solve this problem, based on its capability in precisionly identify pests and diseases of fruit trees and field crops, SenseAgro swiftly applied AI technology for precision identification of the FAW, and adopted the state-of-the-art deep learning technology to build an FAW identification model. At the same time, targeted algorithm optimization was also carried out on the data needed for model training, so as to meet the needs of farmers for field identification as much as possible.

In the process of creating the "Field FAW Identification System", SenseAgro, taking into account the actual application scenarios of farmers, collected a huge number of real image data in the fields with crop protection experts and technical experts of FMC, and established a database covering the whole-life cycle of the pest from egg masses, larvae to adults and the corresponding symptoms of damage to crops, therefore, it is possible to precisely identify FAW larvae or adults and their age stage based on the features of pest and symptoms of damage. At the same time, combined with FMC's abundant application technologies in field control of the FAW, the system can help farmers know well the information of FAW in time and recommend appropriate and prompt control programs, so as to faciliate early detection for early pesticide application and early treatment, and effectively help farmers protect crop safety and crop yield.

Intelligently recommended solutions and control measures help improve the FAW control efficiency

To address the issues of low efficiency in FAW control by farmers and their lack of common knowledge about the pest, an AI identification tool based on WeChat applet was iteratively developed quickly by the algorithm development team of the "Field FAW Identification System" by using AI technology to adapt to application scenarios on strength of a deep understanding of growers' needs and application scenarios. The applet is a field intelligent identification tool to satisfy farmers' needs for pest diagnosis. The attempts and efforts of both SenseAgro and FMC enable farmers to experience high convenience and efficiency brought about by artificial intelligence on a mobile phone.

The "Field FAW Identification System" integrating field application, identification of age stage of FAW larvae and solution recommendation is officially launched in China on July 3, 2020. For FAW control, it must start from identification of the pest. In the whole process of FAW control, it is vital to know the situation of the pest as soon as possible. In this regard, FMC and SenseAgro’s crop protection expert team innovatively used the deep learning technology to distinguish the age stage of FAW larvae and recommend the right pesticide application programs in consideration of the application effect of field pest control and the possible mix-up of Noctuidae with FAW. 

Farmers can identify pests through an applet named "SenseAgro | FAW and Pests & Diseases Identification" by taking photos or uploading photos of the FAW, and obtain corresponding application solutions and the knowledge about the pest intelligently recommended by the applet according to the identification results. The applet can help farmers make sure whether a field is endangered by the FAW in advance and effectively help farmers solve the problems of "early detection" and "early control" of the FAW.

Future cooperation can be expected between FMC and SenseAgro in further development and improvement

Digital agriculture is one of the trends of agricultural development in the future. FMC is cooperating with SenseAgro to work on the application of digital technology in FAW control while continuing to explore, upgrade and optimize the product and add more functions to it to offer more application scenarios to mobile phone users with AI technology.