Smart Agriculture: IoT-based Greenhouse Monitoring System

Abstract

The term "smart agriculture" refers to a management concept that is centered on industrial agriculture. It makes use of cutting-edge technologies such as big data, cloud, and the Internet of Things to monitor, automate, and evaluate agricultural operations. Smart agriculture is a management concept. Software and sensors are used in smart agriculture, also known as precision agriculture. Smart agriculture is managed by the software. This pa-per proposes the development of low-cost environment parameters and electrical quantities monitoring device. According to the findings, such Internet of Things devices are suited for digitizing the relevant data in order to ac-quire information that may help farmers make optimal changes.

Automated Recognition Systems: Theoretical and Practical Implementation of Active Learning for Extracting Knowledge in Image-based Transfer Learning of Living Organisms

“…Based on these findings, the research methodology was designed to focus on CNN-based active learning for the recognition and classification of vine leaves. The theoretical approaches discussed in specialized literature [56], [28], [13] were considered during the methodology design phase. These approaches suggest that, when employing transfer learning, it is important to address three key questions: What is transferred?…”

“…CNNs consist of convolution layers for feature learning, max-pooling layers for dimensionality reduction, and fully connected layers for classification [65]. This research is frequently applied to image recognition and classification tasks [13], as CNNs can learn features, achieve human-level recognition, and adapt to new tasks through re-learning and transfer learning, thereby generating domain knowledge. Our paper aligns with the latest current, where a class of machine learning algorithms utilizes multiple layers to extract significant features gradually from input data [2].…”

“…Initially, the field of artificial intelligence aimed to develop "thinking machines" with general intelligence similar to humans [57]. However, this goal was deemed too ambitious and challenging to achieve directly, leading to a shift in focus towards developing AI systems capable of managing specific tasks in narrow but diverse fields [59], such as agriculture [13], healthcare [84], education [65], retail [22], and the service industry [62]. The ultimate aim is to create high-performance systems capable of processing vast amounts of information, surpassing the processing capacity of the human brain [54].…”

Smart Agriculture in the Digital Age: A Comprehensive IoT-Driven Greenhouse Monitoring System

“…A new architecture (Fig. 2) compared to the one presented in [31] was designed, in the sense that everything was adopted for the recorded data to be transmitted over long distances, using LoRaWAN. We are talking about 7 greenhouses located on an area of 3.64 km 2 .…”

“…This solution is now enabled by LoRaWAN data transmission, primarily because some of the greenhouses are situated far away from the offices. Initially, in [31] a first version of the device was presented, that transmitted the recorded data via WiFi technology. As the project continued to expand and the interest of the participants increased, a need arose for a device that could transmit data over large geographical area with low energy consumption.…”