Akademik Veri Yönetim Sistemi
JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY, cilt.40, sa.3, ss.1559-1572, 2025 (SCI-Expanded, Scopus, TRDizin)
In environmental monitoring studies, monitoring image data and transferring the image to a specific center point for this purpose is very important. Using the LoRa network to perform environmental monitoring through image transmission in geographical areas outside the cellular network coverage area poses difficulties due to the large size of the image data and the high energy requirement for transmission. In this study, rather than transmitting each image; It is aimed to overcome this difficulty in an optimum way by first analyzing the images and transmitting the risky results via LoRa. The study is based on the fact that when a foreign face is detected by performing face recognition with the ResNet model, the LoRa node in the sending position transmits this data and the face image to the central LoRa node. Before the image is transmitted, pixel rounding technique is applied, and then it is compressed with the Huffman Algorithm and divided into packets. After transmission, the packets are combined and opened with the same technique. The study is important in terms of ensuring environmental security in environments where there is no internet access, with LoRa technology, which requires low energy and has the ability to communicate over kilometers long distances. In our study, it has been demonstrated, supported by real environment experiments, that LoRa can be used in environmental monitoring studies aiming to ensure environmental security and control for areas where there is no internet access at long distances, except for risky applications that cannot tolerate data loss. The results of the study show that image transmission with LoRa is possible, although it has difficulties due to its large size.