糯稻虫情测报灯分析数据

通过分析虫害类别，计算虫害数量、占比，实现糯稻农田病虫害状况自动检测，可以了解不同种类昆虫的数量和分布情况。一方面，结合相关的农药知识和防治经验，可以根据不同的虫害情况和程度，精确计算所需的农药用量和配比，实现农药的精准配比和高效灭虫害。即可以避免过量使用农药，减少对环境的影响，同时节约成本和资源。另一方面，监测结果和统计数据可以为农业管理者和农药施用人员提供参考，帮助他们做出更科学和准确的防治决策。根据不同的虫害情况和发展趋势，可以及时调整防治策略，采取有效的措施，高效灭除虫害，保护糯稻作物的生长，为农业现代化提供服务。1. 数据来源
在糯稻农田安装智能虫情测报灯，自动虫情测报灯可对昆虫的发生、发展进行实时自动拍照、实现图像采集。采集数据字段包括创建时间、虫情图片表的id。
2. 数据处理
对于收集到的图像数据，使用基于深度学习，目标检测算法YOLOv5的虫害识别模型，模型输出每个虫害目标的类别标签，识别图像中的虫害名称。在识别结果中，将目标标记为一个矩形框，然后计算这些框的数量，统计每个类别虫害的数量。
每天收集的多张图像数据，统计稻纵卷叶螟、飞虱科、玉米螟、水龟虫的数量，其它虫害数量以及虫害总数量，每个虫害数量分别除以总数量得到稻纵卷叶螟、飞虱科、玉米螟、水龟虫以及其它虫害占比。
3. 数据应用
分析虫害类别，计算虫害数量、占比，可以监测糯稻农田病虫害状况，为防治决策提供参考，为农业现代化提供服务。

This dataset enables automatic detection of pest and disease conditions in glutinous rice farmlands by analyzing pest categories and calculating pest quantities and their proportions, allowing for the acquisition of the quantities and distribution patterns of different insect species. On one hand, combined with relevant pesticide knowledge and pest control experience, precise calculation of the required pesticide dosage and formulation can be performed based on different pest situations and severity levels, achieving precise pesticide formulation and efficient pest control. This not only avoids excessive pesticide use, reduces environmental impacts, but also saves costs and resources. On the other hand, monitoring results and statistical data can provide references for agricultural managers and pesticide applicators, helping them make more scientific and accurate control decisions. By adjusting control strategies in a timely manner based on different pest situations and development trends, effective measures can be taken to efficiently eliminate pests, protect the growth of glutinous rice crops, and serve agricultural modernization.

1. Data Source
Intelligent pest situation monitoring lamps are installed in glutinous rice farmlands. These automatic pest situation monitoring lamps can conduct real-time automatic photography and image collection for the occurrence and development of insects. The collected data fields include creation time and the ID of the pest situation image table.

2. Data Processing
For the collected image data, a pest recognition model based on the deep learning-based object detection algorithm YOLOv5 is used. The model outputs the category labels of each pest target and identifies the pest names in the images. In the recognition results, each target is marked with a bounding box, then the number of these boxes is counted, and the quantity of each category of pests is tallied. For multiple image datasets collected each day, the quantities of *Cnaphalocrocis medinalis*, Delphacidae, *Ostrinia furnacalis*, Hydrophilidae, other pests, and the total number of pests are counted. The proportion of each pest category is obtained by dividing the quantity of each pest by the total pest quantity, including the proportions of *Cnaphalocrocis medinalis*, Delphacidae, *Ostrinia furnacalis*, Hydrophilidae, and other pests.

3. Data Application
By analyzing pest categories and calculating pest quantities and their proportions, the pest and disease conditions in glutinous rice farmlands can be monitored, providing references for pest control decisions and serving agricultural modernization.

搜集汇总

数据集介绍

特点

糯稻虫情测报灯分析数据包含270条记录，通过智能虫情测报灯采集图像数据，利用YOLOv5算法识别虫害类别并统计数量及占比，用于糯稻农田病虫害监测和精准防治决策。