退出

页面顶部背景图片
中文会员 开通

中文文献批量获取

外文会员 开通

外文文献批量获取

  • 首页> 外文期刊>Infrared physics and technology >Detection of pre-symptomatic rose powdery-mildew and gray-mold diseases based on thermal vision
    【24h】

    Detection of pre-symptomatic rose powdery-mildew and gray-mold diseases based on thermal vision

    机译:基于热视力的症状前玫瑰粉状霉菌和灰霉病的检测

    获取原文
    获取原文并翻译 | 示例
           

    摘要

    Highlights ? Potential of thermal imaging in early detection of plant disease was assessed. ? A pre-symptomatic temperature decrease was seen in powdery mildew infected leaves. ? B. cinerea fungus decreased the petals temperature on the first formed lesions. ? Best performance in classifying the infected plants was obtained as early as 2days after inoculation. ? Best classification performance obtained using ANFIS in comparison to RBFNN, MLFNN, and SVM. Abstract Roses are the most important plants in ornamental horticulture. Roses are susceptible to a number of phytopathogenic diseases. Among the most serious diseases of rose, powdery mildew (Podosphaera pannosa var. rosae) and gray mold (Botrytis cinerea) are widespread which require considerable attention. In this study, the potential of implementing thermal imaging to detect the pre-symptomatic appearance of these fungal diseases was investigated. Effects of powdery mildew and gray mold diseases on rose plants (Rosa hybrida L.) were examined by two experiments conducted in a growth chamber. To classify the healthy and infected plants, feature selection was carried out and the best extracted thermal features with the largest linguistic hedge values were chosen. Two neuro-fuzzy classifiers were trained to distinguish between the healthy and infected plants. Best estimation rates of 92.55% and 92.3% were achieved in training and testing the classifier with 8 clusters in order to identify the leaves infected with powdery mildew. In addition, the best estimation rates of 97.5% and 92.59% were achieved in training and testing the classifier with 4 clusters to identify the gray mold disease on flowers. Performance of the designed neuro-fuzzy classifiers were evaluated with the thermal images captured using an automatic imaging setup. Best correct estimation rates of 69% and 80% were achieved (on the second day post-inoculation) for pre-symptomatic appearance detection of powdery mildew and gray mold diseases, respectively. ]]>
    机译:<![cdata [ 亮点 评估早期检测植物病的热成像的潜力。 前症状温度降低在粉状霉菌感染的叶子中看到。 b。碎片真菌在第一个形成的病变上减少了花瓣温度。 在接种后的2 时,获得了分类的最佳性能。 / ce:para> 与RBFNN,MLFNN和SVM相比,使用ANFI获得的最佳分类性能。 抽象 玫瑰是观赏园艺中最重要的植物。玫瑰易患许多植物疗法疾病。在玫瑰最严重的疾病中,白粉病( podosphaera pandosa var。rosae )和灰色模具( botrytis cinerea )是广泛的相当大的关注。在这项研究中,研究了实施热成像以检测这些真菌疾病的前症状外观的可能性。经增长室进行的两次实验检查了粉末状霉菌和灰霉病患者对玫瑰植物的影响( rosa hybrida l。为了分类健康和感染的植物,进行特征选择,选择了最佳提取的热特征,具有最大的语言对冲值。两个神经模糊的分类器培训以区分健康和受感染的植物。在训练中实现了92.55%和92.3%的最佳估计率和92.3%,用8个簇测试分类器,以识别用粉状霉菌感染的叶子。此外,在培训和测试分类器中,最佳估计率为97.5%和92.59%,具有4个集群,以识别花上的灰色霉菌病。使用自动成像设置捕获的热图像评估设计的神经模糊分类器的性能。最佳正确估算率为69%和80%,分别实现了粉末状霉菌和灰霉病患者前症状外观检测的69%和80%。 ]]>

    著录项

    相似文献

    • 外文文献
    • 中文文献
    • 专利
    获取原文

    客服邮箱:kefu@zhangqiaokeyan.com

    京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
    • 客服微信

    • 服务号