Artificial intelligence - enabled soft sensor and internet of things for sustainable agriculture using ensemble deep learning architecture
参考中译:使用集成深度学习架构实现可持续农业的人工智能软测量和物联网


          

刊名：Computers and Electrical Engineering
作者：Anupong Wongchai（Department of Agricultural Economy and Development, Faculty of Agriculture, Chiang Mai University）
Surendra Kumar Shukla（Department of Computer Science & Engineering, Graphic Era Deemed to be University）
Mohammed Altaf Ahmed（Department of Computer Engineering, College of Computer Engineering & Sciences, Prince Sattam Bin Abdulaziz University）
Ulaganathan Sakthi（Department of Computer Science and Engineering, Saveetha School of Engineering, SIMATS）
Mukta Jagdish（Department of Information Technology, Vardhaman College of Engineering (Autonomous)）
Ravi kumar（Department of Electronics and Communication Engineering, Jaypee University of Engineering and Technology）
刊号：738C0073
ISSN：0045-7906
出版年：2022
年卷期：2022, vol.102
页码：108128-1--108128-15
总页数：15
分类号：TP39
关键词：Agriculture；Predictive maintenance；CPS；Soft sensors；Deep learning；Feature representation；Classification
参考中译：农业；预测性维护；CPS；软测量；深度学习；特征表示；分类
语种：eng
文摘：IoT (Internet of things) and Artificial Intelligence (AI), as well as other advanced computing technologies, have long been used in agriculture.AI-enabled sensors function as smart sensors and IoT has made various types of sensor-based equipment in the field of agriculture. This research proposes novel techniques in AI technique based soft sensor integrated with remote sensing model using deep learning architectures. The input has been pre-processed to recognize the missing value, data cleaning and noise removal from the image which is collected from the agricultural land. The feature representation has been carried out usingweight-optimized neural network with maximum likelihood (WONN_ML). After representing the features, classification process has been carried out using ensemble architecture of stacked auto-encoder and kernel-based convolution network (SAE_KCN). The experimental results have been done for various crops in terms of computational time of 56%, accuracy 98%, precision of 85.5%, recall of 89.9% and F-1 score of 86% by proposed technique.
参考中译：物联网(IoT)和人工智能(AI)等先进计算技术在农业中的应用由来已久。AI使能的传感器起到智能传感器的作用，物联网在农业领域制造了各种基于传感器的设备。本研究提出了基于人工智能技术的软测量与基于深度学习的遥感模型集成的新技术。对采集到的农用地图像进行了缺失值识别、数据清洗和去噪处理。特征表示采用加权最大似然神经网络(WONN_ML)。在描述特征的基础上，利用堆叠式自动编码器和基于核的卷积网络的集成结构(SAE_KCN)进行分类处理。对不同作物的实验结果表明，该方法的计算时间为56%，准确率为98%，准确率为85.5%，召回率为89.9%，F-1评分为86%。



国家科技图书文献中心  全球文献资源网  京ICP备05055788号-26  京公网安备11010202008970号  机械工业信息研究院 2018-2024