Conventional lifestyles and working methods are motivated by rising trends and new technologies to form smart cities. Distributed chains are growing to computerize with a highly complicated framework. It is becoming an essential source of additional advantages in today's digital and smart city. Customers or users of smart cities are highly concerned about the quality of food and farm products. On the other hand, tracking the source information is challenging and, preserving its provenance throughout the supply chain network. Traditional supply chain networks are centralized, where a third party does the trading. These centralized systems lack of quality transparency and distribution accountability. Therefore, we provide a promising approach with a unique strategy for converting conventional agriculture to smart agriculture, concerning smart cities, considering blockchain technology and the Internet of Things (IoT) features. The blockchain technique is responsible for maintaining the immutability of the existing network. Still, it faces problems resolving fundamental issues in the distribution chain, like the trustworthiness of the different involved members, transaction responsibility, and real-time tracking. As a result, a trustworthy organization with traceability, conviction, confirmed transportation, and transparent mechanisms in the e-farm supply chain is proposed. Ethereum is used to deploy the proposed smart contract. Testing with the Hyperledger caliper measuring tool shows performance evaluation of several factors such as latency, throughput, resource consumption transaction per second, and so on are assessed. The findings support the effectiveness of the recommended strategy.
The combination of edge computing and deep learning helps make intelligent edge devices that can make several conditional decisions using comparatively secured and fast machine learning algorithms. An automated car that acts as the data-source node of an intelligent Internet of vehicles or IoV system is one of these examples. Our motivation is to obtain more accurate and rapid object detection using the intelligent cameras of a smart car. The competent supervision camera of the smart automobile model utilizes multimedia data for real-time automation in real-time threat detection. The corresponding comprehensive network combines cooperative multimedia data processing, Internet of Things (IoT) fact handling, validation, computation, precise detection, and decision making. These actions confront real-time delays during data offloading to the cloud and synchronizing with the other nodes. The proposed model follows a cooperative machine learning technique, distributes the computational load by slicing real-time object data among analogous intelligent Internet of Things nodes, and parallel vision processing between connective edge clusters. As a result, the system increases the computational rate and improves accuracy through responsible resource utilization and active–passive learning. We achieved low latency and higher accuracy for object identification through real-time multimedia data objectification.