A dataset of observed phenology images of 11 crops including sorghum in China

The change of phenology of crops directly affects the accumulation, quality and yield of photosynthetic products. Based on the historical photos data of 642 national agrometeorological observation stations and 69 national agrometeorological experimental stations and the photos data taken by professional agrometeorological observers, 1376 high-quality photos that can objectively reflect the structure and morphological characteristics of crops were selected from tens of thousands of photos. Through historical photo data collection, missing photos supplement, screening and sorting, multiple rounds of expert identification, data quality control and quality evaluation and other links, and in strict accordance with the standard of crop phenology specified in the “specification for agrometeorological observation (Volume I)”, 179 photos were selected to build the dataset. It includes 11 sub folders, respectively storing 27 photos of sorghum, 17 photos of foxtail millet, 12 photos of sweet potato, 14 photos of potato, 24 photos of peanut, 13 photos of sesame, 17 photos of sunflower, 9 photos of sugarcane, 17 photos of flax, 9 photos of sugar beet and 20 photos of tobacco, a total of 179 photos of key development stages of crops, with a total size of 516.89 MB. The dataset is classified according to the characteristics of crops and their phenological phases, with their growth and development process as the main line, so as to directly reflect the morphological structure characteristics of crops at different growth stages. The photo dataset can comprehensively display the technical standards and methods of agrometeorological observation of crops in the phenological phase. It can help agrometeorological observers to grasp the manual observation technology quickly, accurately and systematically in a short time, and improve the operational ability of agrometeorological observation and the quality of observation data. It can provide reliable data support for the research of automatic identification algorithm of crop phenological phase, agricultural production strategies to deal with climate change, and has important reference value for the research and technical training of agrometeorological manual observation technology, cultivation and management of crops, related education and popular science curriculum development.作物物候（crop phenology）的变化直接影响光合产物的积累、品质与产量。本数据集依托全国642个农业气象观测站、69个农业气象试验站的历史影像数据，以及专业农业气象观测人员拍摄的影像资料，从数万张影像中筛选出1376张可客观反映作物结构与形态特征的高质量影像。历经历史影像数据采集、缺失影像补全、筛选整理、多轮专家鉴定、数据质量管控与质量评估等多个环节，并严格遵循《农业气象观测规范（第一卷）》中规定的作物物候标准，最终遴选出179张影像构建本数据集。数据集包含11个子文件夹，分别存储高粱（sorghum）影像27张、谷子（foxtail millet）影像17张、甘薯（sweet potato）影像12张、马铃薯（potato）影像14张、花生（peanut）影像24张、芝麻（sesame）影像13张、向日葵（sunflower）影像17张、甘蔗（sugarcane）影像9张、胡麻（flax）影像17张、甜菜（sugar beet）影像9张，以及烟草（tobacco）影像20张，共计179张作物关键生育期影像，总容量达516.89 MB。本数据集按照作物特征及其物候相（crop phenological phases）进行分类，以作物生长发育进程为主线，可直观展现作物不同生育阶段的形态结构特征。该影像数据集可全面展示作物物候期农业气象观测的技术标准与实施方法，能够帮助农业气象观测人员在短时间内快速、准确、系统地掌握人工观测技术，提升农业气象观测业务能力与观测数据质量。本数据集可为作物物候期自动识别算法研究、应对气候变化的农业生产策略研究提供可靠的数据支撑，对农业气象人工观测技术的研究与技能培训、作物栽培管理、相关科普教育及课程开发等工作均具有重要参考价值。

搜集汇总

背景与挑战

背景概述

该数据集是一个包含高粱、谷子、甘薯等11种作物在中国关键发育阶段的观测照片集合，共179张高质量图像，总大小为516.89 MB。数据来源于全国农业气象观测站和历史照片，经过严格筛选和质量控制，旨在支持农业气象观测技术培训、作物物候期自动识别算法研究以及应对气候变化的农业生产策略分析。