Figure 1 : Object processor of the Intelligent Internet of Things ( I-IoT ) connects IoT devices or applications to the processing center on the cloud . ( Source : IEEE Communications Magazine , vol . 56 , no . 10 , pp . 114-120 ).
AGRICULTURAL I-IoT
Applying analytics on agricultural data streams is crucial to synthesise hidden knowledge and useful information in improving operational efficiency and increasing productivity ; albeit , not a straightforward and easy task . For instance , image-processing techniques have been used extensively to solve problems that often involve detailed inspection of a large amount of materials or objects , or simply , data , covering various purposes ranging from detection of pest and disease in leaf , stem , and fruit , quality monitoring of fruits , and weed detection . Recently , the combination of image processing and IoT is carried out to improve agricultural produce quality and reduce crop failure . Common examples include using drones to capture aerial images at regular intervals as well as using vision-based IoT devices to monitor environmental factors . Unlike applications that utilise the traditional rule-based algorithm or inferences approaches , more complex agricultural problems require new technologies , algorithms and infrastructures . Fortunately , recent progresses in both fast computing and advanced machine learning techniques are opening doors for big data analytics and knowledge extraction suitable for IoT applications . Meanwhile , the evolution of communication networks towards extended coverage , higher throughput , lower latency and massive bandwidth , such as 5G cellular network , paves the way for better transmission of agricultural data . In view of these recent developments , the novel paradigm that unites the Internet , things and intelligence , namely Intelligent Internet of Things ( I-IoT ) is introduced .
Additional components that enable the transformation of IoT into I-IoT are 1 ) object
13