Automated per-plot leaf-level imaging and analysis for small plot arable field trials

挑战粮农组织估计，到2050年，粮食产量需要增加70%，以实现全球粮食安全。开发新的作物品种、农用化学品和作物管理技术是提高粮食生产力和减少环境影响的关键。它们依赖于多阶段测试;最初是在温室中进行实验室测试，然后在多种土壤类型和气候下进行实地试验。4家最大的农业公司的研发预算合计超过50亿美元：* 在英国; NIAB运行150，000多个地块，Agrii，40，000多个 * 全球; UPL运行1500个多地块试验，Syngenta运行114个研发中心试验计划迫切需要创新，因为大多数监测仍然是视觉，手动和亲自完成的。 无人机技术和人工智能图像分析可以提供更频繁的监控，与其他技术相比节省了大量时间和成本（速度最多快30倍），加上统一的数据捕获/分析，从而加快产品开发速度并改善环境状况。微型中小企业业务，Drone Ag将开发一种新的试验图成像和分析系统，在几分钟内提供独特的每图，极高分辨率的分析。通过这项技术，Drone Ag将实现：* 劳动力技能提升;从人工试用评估中解放出来。通过节省时间/成本，快速进行化学和育种迭代。*数据供应，为以前无法以如此低的成本/高的体积比重复收集统一数据的公司提供新的人工智能模型。由于这将用更高效，更详细的系统取代当前的做法，因此该解决方案具有高度创新性，颠覆性，并将受到高度追捧。我们将在整个项目过程中及以后利用关键行业利益相关者的反馈，以确保解决方案满足他们的要求。该项目将为最终用户和更广泛的供应链带来经济利益。我将把它转化为对整个人口的更广泛利益，包括更健康和更清洁的空气，水和食物。

CHALLENGE The FAO estimates that food production needs to increase 70% by 2050 to achieve global food security. Development of new crop varieties, agrochemicals and crop management techniques are key to increasing food productivity and reducing environmental impact. They rely on multi-phase testing; initially lab based, in greenhouses, then in field trials across multiple soil types and climates. The 4 largest Ag companies have a combined R&D budget of over $5bn:* In the UK; NIAB run 150,000+ plots, Agrii, 40,000+* Worldwide; UPL run 1500 multi-plot trials, Syngenta run 114 R&D sitesTrials programmes are in desperate need of innovation, As most monitoring is still done visually, manually, and in-person. SOLUTION Drone tech and AI image analysis can provide more frequent monitoring with notable time and cost savings over other techniques (up to 30x faster) plus uniform data capture/analysis for faster product development and better environmental profiles. Micro-SME business, Drone Ag will develop a new trials plot imaging and analysis system to deliver unique per-plot, extreme high resolution analytics within minutes. Through this technology, Drone Ag will enable:* Workforce upskilling; freed from manual trial evaluation by eye.* Rapid chemical and breeding iteration through time/cost saving.* Data provision, feeding new AI models for companies who previously could not repeatedly gather uniform data at such a low cost/high volume ratio.Since this would replace current practice with a more efficient, detailed system, this solution is highly innovative, disruptive and would be highly sought-after. We will be making use of feedback from key industry stakeholders throughout the project and beyond, to ensure the solution meets their requirements. The project will deliver economic benefits to and end-users, and the wider supply-chain.I would translate to wider benefits for the population as a whole including healthier and cleaner air, water and food.