Cucumber Disease Recognition Dataset

(1) Crop disease is a widespread problem in the productivity and quality of agricultural production. It adversely affects the quality of crops. The cucumber is a frequently grown creeping vine plant that has few calories but is high in water and several vital vitamins and minerals. Due to the non-biological circumstances, cucumber diseases will adversely harm the yield and quality of cucumber and cause heavy economic losses to farmers. The traditional diagnosis of crop diseases is often time-consuming, laborious, ineffective, and subjective. (2) In the recent era, computer vision approaches are very promising for handling these kinds of classification and detection tasks. (3) To develop machine vision-based algorithms, a major cucumber dataset is illustrated containing eight types of cucumber classes, namely Anthracnose, Bacterial Wilt, Belly Rot, Downy Mildew, Pythium Fruit Rot, Gummy Stem Blight, Fresh leaves, and Fresh cucumber. Cucumber disease classifications are done with the cooperation of an expert from an agricultural institute. (4) A total of 1280 images of cucumbers are collected from real fields. Then from these original images, a total of 6400 augmented images are produced using flipping, shearing, zooming, and rotation techniques to increase the data number.

(1) 作物病害是影响农业生产生产力与品质的普遍问题，会对作物品质造成负面影响。黄瓜是广泛种植的蔓生藤本作物，热量较低但水分与多种关键维生素、矿物质含量丰富。非生物环境因素诱发的黄瓜病害，会严重损害黄瓜的产量与品质，给种植户造成巨额经济损失。传统作物病害诊断方法往往耗时耗力、效率低下且主观性较强。
(2) 近年来，计算机视觉技术在处理此类分类与检测任务中展现出卓越的应用潜力。
(3) 为研发基于机器视觉的病害识别算法，本研究构建了一套大型黄瓜数据集，涵盖8类黄瓜样本类别，分别为炭疽病（Anthracnose）、细菌性萎蔫病（Bacterial Wilt）、果腐病（Belly Rot）、霜霉病（Downy Mildew）、腐霉果腐病（Pythium Fruit Rot）、瓜类蔓枯病（Gummy Stem Blight）、健康叶片（Fresh leaves）以及健康黄瓜（Fresh cucumber）。本数据集的黄瓜病害分类标注工作由农业科研机构专家协助完成。
(4) 研究团队从大田实景中采集了1280张原始黄瓜图像，随后通过翻转、剪切、缩放及旋转等数据增强技术，从原始图像中生成共计6400张增强图像，以扩充数据集规模。

搜集汇总

背景与挑战

背景概述

该数据集专注于黄瓜疾病识别，包含8个类别，涵盖7种常见疾病（如炭疽病、细菌性枯萎病等）以及健康叶片和黄瓜，用于支持计算机视觉分类任务。数据集由1280张原始田间图像组成，并通过数据增强技术扩展到6400张图像，以提高模型训练效果，所有分类均与农业专家合作完成。