Urban change detection is an important part of sustainable urban planning, regional development, and socio-economic analysis, especially in regions with limited access to economic and demographic statistical data. The goal of this research is to create a strategy that enables the extraction of indicators from large-scale orthoimages of different resolution with practically acceptable accuracy after a short training process. Remote sensing data can be used to detect changes in number of buildings, forest areas, and other landscape objects. In this paper, aerial images of a digital raster orthophoto map at scale 1:10,000 of the Republic of Lithuania (ORT10LT) of three periods (2009–2010, 2012–2013, 2015–2017) were analyzed. Because of the developing technologies, the quality of the images differs significantly and should be taken into account while preparing the dataset for training the semantic segmentation model DeepLabv3 with a ResNet50 backbone. In the data preparation step, normalization techniques were used to ensure stability of image quality and contrast. Focal loss for the training metric was selected to deal with the misbalanced dataset. The suggested model training process is based on the transfer learning technique and combines using a model with weights pretrained in ImageNet with learning on coarse and fine-tuning datasets. The coarse dataset consists of images with classes generated automatically from Open Street Map (OSM) data and the fine-tuning dataset was created by manually reviewing the images to ensure that the objects in images match the labels. To highlight the benefits of transfer learning, six different models were trained by combining different steps of the suggested model training process. It is demonstrated that using pretrained weights results in improved performance of the model and the best performance was demonstrated by the model which includes all three steps of the training process (pretrained weights, training on coarse and fine-tuning datasets). Finally, the results obtained with the created machine learning model enable the implementation of different approaches to detect, analyze, and interpret urban changes for policymakers and investors on different levels on a local map, grid, or municipality level.

Automatic climate management enables us to reduce repetitive work and share knowledge of different experts. An artificial intelligence-based layer to manage climate in white button mushroom growing hall was presented in this article. It combines visual data, climate data collected by sensors, and technologists’ actions taken to manage climate in the mushroom growing hall. The layer employs visual data analysis methods (morphological analysis, Fourier analysis, convolutional neural networks) to extract indicators, such as the percentage of mycelium coverage and number of pins of different size per area unit. These indicators are used to generate time series that represent the dynamics of the mushroom growing process. The incorporation of time synchronized indicators obtained from visual data with monitored climate indicators and technologists’ actions allows for the application of a supervised learning decision making model to automatically define necessary climate changes. Whereas managed climate parameters and visual indicators depend on the mushroom production stage, three different models were created to correspond the incubation, shock, and fruiting stage of the mushroom production process (using decision trees, K-nearest neighbors’ method). An analysis of the results showed that trends of the selected visual indicators remain similar during different cultivations. Thus, the created decision-making models allow for the definition of the majority of the cases in which the climate change or transition between the growing stages is needed.

Hydromorphology of rivers assessed through direct measurements is a time-consuming and relatively expensive procedure. The rapid development of unmanned aerial vehicles and machine learning (ML) technologies enables the usage of aerial images to determine hydromorphological units (HMUs) automatically. The application of various direct and indirect data sources and their combinations for the determination of river HMUs from aerial images was the main aim of this research. Aerial images with and without the Sobel filter, a layer of boulders identified using Yolov5x6, and a layer of direct measurements of depth and streamflow velocity were used as data sources. Three ML models were constructed for the cases if one, two, or three data sources were used. The ML models for HMU segmentation were constructed of MobileNetV2 pre-trained on ImageNet data for the feature extraction part and U-net for the segmentation part. The stratified K-fold cross-validation with five folds was carried out to evaluate the performance of the model due to the limited dataset. The analysis of the ML results showed that the measured metrics of segmentation using direct measurements were close to the ones of the model trained only on the combination of boulder layer and aerial images with the Sobel filter. The obtained results demonstrated the potential of the applied approach for the determination of HMUs only from the aerial images, and provided a basis for further development to increase its accuracy.

The socio-economic development of municipalities is defined by a set of indicators in a period of interest and can be analyzed as a multivariate time series. It is important to know which municipalities have similar socio-economic development trends when recommendations for policy makers are provided or datasets for real estate and insurance price evaluations are expanded. Usually, key indicators are derived from expert experience, however this publication implements a statistical approach to identify key trends. Unsupervised machine learning was performed by employing K-means clusterization and principal component analysis for a dataset of multivariate time series. After 100 runs, the result with minimal summing error was analyzed as the final clusterization. The dataset represented various socio-economic indicators in municipalities of Lithuania in the period from 2006 to 2018. The significant differences were noticed for the indicators of municipalities in the cluster which contained the 4 largest cities of Lithuania, and another one containing 3 districts of the 3 largest cities. A robust approach is proposed in this article, when identifying socio-economic differences between regions where real estate is allocated. For example, the evaluated distance matrix can be used for adjustment coefficients when applying the comparative method for real estate valuation.

Parking lots have wide variety of shapes because of surrounding environment and the objects inside the parking lot, such as trees, manholes, etc. In the case of paving the parking lot, as much area as possible should be covered by the construction vehicle to reduce the need for manual workforce. Thus, the coverage path planning (CPP) problem is formulated. The CPP of the parking lots is a complex problem with constraints regarding various issues, such as dimensions of the construction vehicle and data processing time and resources. A strategy based on convolutional neural networks (CNNs) for the fast estimation of the CPP’s average track length, standard deviation of track lengths, and number of tracks was suggested in this article. Two datasets of different complexity were generated to analyze the suggested approach. The first case represented a simple case with a working polygon constructed out of several rectangles with applied shear and rotation transformations. The second case represented a complex geometry generated out of rectangles and ellipses, narrow construction area, and obstacles. The results were compared with the linear regression models, with the area of the working polygon as an input. For both generated datasets, the strategy to use an approximator to estimate outcomes led to more accurate results compared to the respective linear regression models. The suggested approach enables us to have rough estimates of a large number of geometries in a short period of time and organize the working process, for example, planning construction time and price, choosing the best decomposition of the working polygon, etc.

The non-destructive inspection by ultrasonic immersion imaging of various objects with complex geometric surfaces can be problematic because the effects of refraction have a high impact on the path of propagating acoustic waves. In order to reconstruct an image of sufficient quality, development of a precise surface model of the structure is necessary. In this research, two different methods (the total focusing method (TFM) and plane-wave imaging (PWI) using Delay-And-Sum(DAS) and Multiply-Delay-And-Sum (MDAS) modes) for surface-geometry reconstruction are investigated. The ability of these methods to reconstruct the boundaries of complex surfaces is also presented using probes with less than 10 Mhz center frequency where object surface angles bent radius is comparable with emission wavelength. It is demonstrated that PWI is a more effective method of making surface reconstructions than the TFM due to its simplicity and utilization of lesser computational resources to reconstruct the surfaces of larger objects.