Plant diseases cause great damage in agriculture, resulting in significant yield losses. The recent expansion of deep learning methods has found its application in plant disease detection, offering a robust tool with highly accurate results. The current limitations and shortcomings of existing plant disease detection models are presented and discussed in this paper. Furthermore, a new dataset containing 79,265 images was introduced with the aim to become the largest dataset containing leaf images. Images were taken in various weather conditions, at different angles, and daylight hours with an inconsistent background mimicking practical situations. Two approaches were used to augment the number of images in the dataset: traditional augmentation methods and state-of-the-art style generative adversarial networks. Several experiments were conducted to test the impact of training in a controlled environment and usage in real-life situations to accurately identify plant diseases in a complex background and in various conditions including the detection of multiple diseases in a single leaf. Finally, a novel two-stage architecture of a neural network was proposed for plant disease classification focused on a real environment. The trained model achieved an accuracy of 93.67%.

The aim of this paper is to investigate the effectiveness of combining mobile technologies and sensors to detect harmful particles in the air and address the problem of air pollution caused by traffic and exhaust emissions. The paper contains a systematic literature review of information technology research related to pollution detection in order to point out the main obstacles in the field and propose solutions. Furthermore, the paper presents the development of an Android smartphone-based real-time monitoring system that utilizes an external analog sensor board to acquire and evaluate physical measurements. The proposed system is calibrated for CO gas measurements, and it is compared with a commercial gas analyzer instrument. Results imply that the developed system is capable of detecting concentration levels in the air and that the accuracy is within the range of the industrial device’s accuracy.

An urban environment defines a specific micro-climate which directly affects the quality of life in urbanized areas and often has a negative impact on urban populations. Phenomena like urban heat and surface heat islands are direct products of an urban lifestyle. Urban meteorological networks (UMNs) are a tool that can help to better understand and analyze the current situation and make the right decisions about future urban development. Deployed to monitor and record different objects and their states inside urban areas, UMNs build a long-term meteorological data time series database. The most commonly used systems for achieving this goal include wireless sensor networks (WSNs). This paper presents a combined experience in deploying three different WSN systems. During seven years of research in this field, the authors have recognized the importance of data reliability in data acquisition. More importantly, due to the lack of research addressing the reliability of the data received from WSNs by the core segment of the server (processes used in receiving, validating, parsing, and storing data into a database instance), the received data are used in scientific studies without questioning their reliability. To determine the possibility of shifting information provided by the data measured from sensor networks before it is stored in a desired form of database, this paper proposes a highly reliable socket server model. The model is built with high reliability and performance in mind and it includes three major processes, which use a combination of signals and control messages to pass information about their states. A case study is performed using high-end hardware, running a Linux operating system stressed to its limits. Repetition testing revealed inconsistency in the information provided by the operating system to the application layer, which could lead to the loss of information about short-term and rarely occurring monitored objects. The results lead to the conclusion that there is a clear need for a higher level of data reliability in the process of data acquisition by UMNs. The proposed socket server should fill this gap within the server’s core segment.

The Internet of Things (IoT) encompasses a network of interconnected devices that collect, analyze, and exchange vast amounts of data. However, this connectivity creates opportunities for various types of cyberattacks, making IoT systems vulnerable and potentially leading to the compromise of sensitive information. Therefore, developing effective intrusion detection system (IDS) is one of the key challenges in IoT network security. The aim of this study is to develop a machine learning (ML) model for network traffic classification and attack detection in IoT environments. Through a comparative analysis of different algorithms, the study seeks to identify the model with the best performance, which could serve as a foundation for efficient IDS solutions tailored to the specific characteristics of IoT networks. The RT-IoT2022 dataset was used for experimental analysis, providing realistic framework for testing ML models, including k-nearest neighbors, Random Forest, XGBoost, multilayer perceptron, and various 1D convolutional neural network architectures. The study examines preprocessing techniques, focusing on dimensionality reduction (principal component analysis, variance inflation factor, Pearson’s test), outlier detection (interquartile range, Z -score, Isolation Forest), and transformation methods (Box–Cox, RobustScaler, Winsorization). Based on the results of the experiment, the most effective model and preprocessing technique were proposed.

Image processing in biomedicine is a very broad field, which includes both medical and technical significance. The aim of this paper is to investigate the current trends in the domain of application of U-Net architecture in the period from 2018 to 2023. The PRISMA framework was used for the systematic literature review, and 4 research questions were asked. For the most part, U-Net architectures are used that can process complex high-resolution images in the fastest way in the context of semantic segmentation. Previous work in image processing has focused on overcoming problems such as the complexity of different architectures, image loss, image resolution, and quality, as well as the size of datasets and noise reduction. The most frequently used groups of datasets are BraTS, Data Science Bowl, and ISIC Challenge. The best general Dice score was obtained for the LUNA16, VESSEL12, and Kaggle Lung datasets with 0.98. It is concluded that the application of the U-Net network is growing, with a focus on solving specific challenges in the context of a certain modality and segment of biomedicine.