Citrus Pest Benchmark (CPB)
收藏资源简介:
本研究创建了名为Citrus Pest Benchmark (CPB)的数据集,由复杂数据推理实验室(RECOD)开发。该数据集包含10,816张图像,分为七个类别,主要用于农业领域的柑橘害虫识别。数据集中的图像通过移动设备结合放大镜拍摄,涵盖了六种螨类害虫和一种负样本类别。创建过程涉及在巴西的São José农场进行定期检查,采集害虫图像。该数据集旨在通过自动化技术辅助集成害虫管理(IPM),提高农业害虫检测的效率和准确性。
This study developed a dataset named Citrus Pest Benchmark (CPB), which was constructed by the Complex Data Reasoning Laboratory (RECOD). This dataset contains 10,816 images grouped into seven classes, and is primarily intended for citrus pest recognition in the agricultural field. All images in the dataset were captured using mobile devices equipped with magnifying glasses, covering six categories of mite pests and one negative sample class. The dataset creation process involved regular inspections at the São José Farm in Brazil to collect pest images. This dataset aims to support Integrated Pest Management (IPM) through automated technologies, thereby improving the efficiency and accuracy of agricultural pest detection.




