Aerial Photographs Research Articles

Digitizing Historical Aerial Images: Evaluation of the Effects of Scanning Quality on Aerial Triangulation and Dense Image Matching

In the last decade, many aerial photographic archives have started to be digitized for multiple purposes, including digital preservation and geoprocessing. This paper analyzes the effects of professional photogrammetric versus consumer-grade scanners on the processing of analog historical aerial photographs. An image block over Warsaw is considered, featuring 38 photographs acquired in 1986 (Wild RC10, Normal Aviogon II lens, 23 × 23 cm format) with a ground sampling distance (GSD) of 4 cm. Aerial triangulation (AT) and dense image matching (DIM) procedures are considered, analyzing how scanning modalities are important in the massive digitization of analog images for georeferencing and 3D product generation. The achieved results show how consumer-grade scanners, unlike more expensive photogrammetric scanners, do not possess adequate recording quality to ensure high accuracy and geometric precision for geoprocessing purposes. However, consumer-grade scanners can be used for time and cost-efficient applications where a partial loss of data quality is not critical.

Utilizing Aerial Photography to Trace the Development of Urbanization in the Chennai District

Chennai, India is growing more urbanized by the day, hence studying its development is essential. We may see the world from a bird's-eye viewpoint thanks to aerial photography, a method for taking pictures of the Earth's surface from an elevated position—typically from an aircraft or drone. Scientific research, land surveying, urban planning, environmental monitoring, and mapping are just a few of the applications for these photos. Using remote sensing techniques, we are able to examine and record changes in Chennai's physical terrain over time, allowing us to trace the city's evolution through these aerial photos. ArcGIS, ERDAS, ENVI, Pix4D, Global Mapper, and QGIS were among the programs we used to process and examine the aerial photo data. Through the help of these software programs, users can extract insightful data and useful information from photos taken by satellites, drones, or airplanes. When compared to satellite images, aerial photography frequently produces imagery with a better spatial resolution, enabling more accurate and thorough study of ground-based objects. Through a comparative analysis of maps produced using aerial photography data throughout time, we have discovered that Chennai is losing open land, green spaces, and water bodies because of its growing urbanization. By closely analysing this trend and taking into account various factors, we can forecast the continued growth of Chennai.

Exponential growth of private coastal infrastructure influenced by geography and race in South Carolina, USA

Homeowners in coastal environments often augment their access to estuarine ecosystems by building private docks on their personal property. Despite the commonality of docks, particularly in the Southeastern United States, few works have investigated their historical development, their distribution across the landscape, or the environmental justice dimensions of this distribution. In this study, we used historic aerial photography to track the abundance and size of docks across six South Carolina counties from the 1950s to 2016. Across our roughly 60-year study period, dock abundance grew by two orders of magnitude, mean length of newly constructed docks doubled, and the cumulative length of docks ballooned from 34 to 560 km. Additionally, we drew on census data interpolated into consistent 2010 tract boundaries to analyze the racial and economic distribution of docks in 1994, 1999, 2011, and 2016. Racial composition, measured as the percentage of a tract’s population that was White, positively correlated with dock abundance in each year. Median household income and dock abundance were only correlated in 2011. Taken together, these metrics indicate the growing desire for direct estuary access, however, that access does not appear to be equally spread across racial groups. Because docks enhance estuarine access and demarcate private property, our study provides longitudinal insights into environmental justice concerns related to disparate private property ownership. We found a persistent correlation between the racial characteristics of an area and dock abundance, strongly indicating that White South Carolinians have had disproportionately greater private water access for the past two decades.

Inferring channel incision in gravel‐bed rivers: Integrating LiDAR data, historical aerial photographs and drone‐based SfM topo‐bathymetry

AbstractChannel incision is an evident trend for river evolution in many European rivers and notably the Western Carpathians, whose former braided and multichannel wandering river system is transforming into a single‐thread channel, but it is often difficult to separate drivers and determine if incision is finished or is still ongoing. To overpass these research gaps, this paper presents an innovative approach to assess the multidecadal incision of the Belá River in the Western Carpathians since 1949 by LiDAR‐based analyses of floodplain surfaces above the river channel dated from historical aerial images. Detailed analyses of ongoing incision were also calculated based on DEM of differences (DoD) using Structure‐from‐Motion (SfM) photogrammetry‐derived topo‐bathymetric models. The study applied the BACI (Before‐After‐Control‐Impact) approach that compared pre‐state (Before), post‐state (After) and reach (Control) that is not affected by potential external effects with degraded (impacted) reach to be able to distinguish the driver effects. Floodplain channel surface analyses indicate the maximum incision up to 4 m and incision rate of 5.7 cm/year that occurred in the most degraded reach. Moreover, cross‐section profiles point to accelerated incision of 24.5 cm/year in the last 10 years (2011–2021) by the propagation of incision upstream. Overall, the net changes from the UAV survey pointed to 22 759 m3 of gravel sediments, constituting outwash from the 1.6 km long channel system (2015–2022) by incision, whereas analyses of historical channel surfaces estimated erosion of 573 303 m3 from impacted reaches between 1949 and 2020. Incision evidence is only observed in the downstream part below the control section due to local drivers (channel regulation, comprising embankment and gravel mining that activated a backward erosion of the system with knickpoint migration upstream). This analysis shows the benefits of combining different sources of data to separate long‐term and ongoing channel responses and the BACI‐approach to better target cause–effect relationships in space and time.

Quantification of the spatial distribution of individual mangrove tree species derived from LiDAR point clouds

AbstractMangrove forests have unquestionably high environmental and ecological value. Mangrove trees are believed to have habitat zonation that is controlled mainly by the relative sea level. However, earlier discussions of mangrove habitats have remained limited in terms of their quality and quantity because of a lack of high-resolution spatial information of microtopography and trees. To clarify mangrove habitability over a wide forest area, we compounded mobile laser scanning (MLS) and aerial laser scanning (ALS) LiDAR dataset of the Miyara River mangrove on Ishigaki Island, Okinawa, Japan. The MLS provided sub-canopy data, while the unmanned aerial vehicle ALS data mainly provided a point cloud of the canopy. We corrected point clouds and combined these data. The results indicated that ALS is unable to reconstruct the microtopography of the dense mangrove area well. Moreover, tree species were not identifiable from the ALS data. However, by applying MLS to the mangrove forest, we obtained high-resolution microtopography and tree information inside the forest, although the measurement area was limited to comparison with ALS. By combining ALS and MLS point clouds, 3D point clouds of the forest were well reconstructed. From these point clouds, a high-resolution digital elevation model was created. Subsequently, we segmented trees individually from composite MLS point clouds and identified each tree species. Consequently, the spatial distribution of thousands of mangrove trees was reconstructed at the Miyara River mouth. The spatial distribution of mangrove tree species together with earlier aerial photographs suggests that mangrove species have been segregated in accordance with changes in their elevation and environment over 40 years. Our findings suggest that the distribution of the species changed sensitively along with dynamic variation of the microtopography.

Design and implementation of an Unmanned Aerial Vehicle (UAV) for image capture in enterprise farming

Unmanned aerial vehicle (UAV), also known as drone is one of the most interesting emerging technologies with a wide range of applications including agriculture, surveillance, security, search and rescue, mapping, farmland surveying, and wildlife conservation, among others. In Nigeria, UAV is gaining popularity in applications such as in social functions for taking both moving and static pictures, military and security, mapping and agriculture. In this work, an UAV was designed, some components produced by 3D printing, coupled and tested with certain level of local content in view. The UAV is a quadcopter equipped with a camera for real-time image capture of farmlands and it can perform autonomous missions by using global positioning system (GPS) waypoints. The weight of the quadcopter was approximately calculated and hence the electrical and mechanical components selected using a standard empirical design table. The thrust to weight ratio was set at 2:1. Autodesk Maya and Inventor software were used to design the frame in two parts: the frame arms on which the propellers are mounted and the central connecting part that links the four arms. The arms are made up of aluminum square pipes while the central connecting part for the arms was printed with a 3D printer (200 x 200 x 200 Wanhao duplicator i3) using PolyEthylene Terephthalate Glycol (PETG) Filament material of 1.75mm standard gauge. Material selection was based on material strength, cost and availability. NCH Debut video and image capture software was used to record live feeds from the UAV onboard camera. Functionality tests for lift, stability, yaw, roll, pitch, loiter, auto landing, return to launch, flight time, altitude/signal, auto mission using GPS waypoints were carried out with the completed UAV on a less windy day to avoid the influence of wind. It was generally observed that the UAV successfully took off the ground, gained stability, flew to over 100m height, captured aerial photographs of the land below it while on flight at the required height and landed safely. Cost saving advantage of 45.28% was achieved when compared to imported equivalent types.

Drones Detection Using a Fusion of RF and Acoustic Features and Deep Neural Networks.

The use of drones has recently gained popularity in a diverse range of applications, such as aerial photography, agriculture, search and rescue operations, the entertainment industry, and more. However, misuse of drone technology can potentially lead to military threats, terrorist acts, as well as privacy and safety breaches. This emphasizes the need for effective and fast remote detection of potentially threatening drones. In this study, we propose a novel approach for automatic drone detection utilizing the usage of both radio frequency communication signals and acoustic signals derived from UAV rotor sounds. In particular, we propose the use of classical and deep machine-learning techniques and the fusion of RF and acoustic features for efficient and accurate drone classification. Distinct types of ML-based classifiers have been examined, including CNN- and RNN-based networks and the classical SVM method. The proposed approach has been evaluated with both frequency and audio features using common drone datasets, demonstrating better accuracy than existing state-of-the-art methods, especially in low SNR scenarios. The results presented in this paper show a classification accuracy of approximately 91% at an SNR ratio of -10 dB using the LSTM network and fused features.

The Spatial Impact of Crowds in the Chilean Social Explosion: A Case Study of Postpolitical Democracy

This paper introduces an innovative methodology for estimating crowd sizes at significant events, particularly focusing on protests. This method combines aerial photography with urban mapping to enhance accuracy in measuring participant numbers. Originating from the analysis of extensive social protests in Chile, this approach was notably applied to what was reported as the largest march, where mainstream media claimed attendance exceeded one million. However, utilizing this new methodology, we recalculated the figure to be 354,167, challenging previous estimations. This research offers a reliable, repeatable tool for quantifying attendance at large gatherings, shedding light on the dynamics of public mobilization and space utilization. It provides a framework for assessing the comparative mobilization strength of various social movements and can extend to other large public events like concerts and sports, helping to gauge public engagement levels. Accurate crowd size estimations can inform policymakers, leading to more informed decisions and fostering a dialogue-based relationship between authorities and the public. By understanding protest scales and impacts, policies can be more aligned with societal needs, promoting a more inclusive and participatory democratic environment. This article, therefore, not only presents a methodological advancement but also underscores its applicability in enhancing democratic governance and public policy formulation.

Impacts of anthropogenic structures in long- and short-term shoreline evolution of Santa Marta Bay (Colombian Caribbean)

Interaction of Anthropic interventions (rigid structures) with coasts is an essential aspect for understanding their geomorphic evolution and incorporating these data into adequate coastal management. This study analyses the short and long-term behavior (1985 to 2019) of shoreline at Santa Marta Bay (Caribbean - Colombia) and their relationship with coastal protection structures. The shoreline variations were analyzed through aerial photographs and satellite images using DSAS tools. The short-term assessment showed initial intense 1985–1991 erosion, with an average retreat speed rate of − 1.6 m·y− 1. It changed gradually since 2003, due to the construction of rigid structures and beach nourishment, decreasing shoreline retreat and even a progradation rate of 0.2 m·y− 1 was observed. The 2009–2019 period, despite recording a positive average value, exhibited a high percentage of erosion profiles. Therefore, in a decadal or long-term analysis (1985–2019), despite engineering works present positive results in the short term, new interventions are required. The coastal erosion is not uniform along the shoreline, because the area is a closed bay with small sediment inputs, the variations in erosion and accretion rates change with human intervention and the installation of new structures. Therefore, quantifying the scale and rate of shoreline changes and correlating them with anthropogenic structures is an essential step in assessing shoreline vulnerability.

Challenges for Sustainable Urban Planning: A Spatiotemporal Analysis of Complex Landslide Risk in a Latin American Megacity

This paper analyzes the spatiotemporal evolution of a complex landslide risk scenario in a Latin American megacity, underscoring the key challenges it poses for sustainable urban planning in such cities. This research draws upon multiple studies commissioned by the mayor’s office of the megacity of Bogota, Colombia, and utilizes aerial photographs and satellite imagery from diverse sensor types. The methodology used considered six spatiotemporal analysis scenarios: rural/natural, mining, urban, landslide risk, stabilization and environmental park, and informal reoccupation. The findings reveal a complex interplay between the megacity’s peripheral areas, which face constraints for human settlement, and their potential for construction material exploitation. This complex relationship was further compounded by weaknesses in planning and controlling peripheral occupations, coupled with a burgeoning demand for developable land in a landslide risk context (landslide area: 73 ha). The analysis scenarios highlighted the predominant use of a reactive urban planning approach that addressed events, changes, or problems after they had occurred, rather than proactively anticipating and preventing potential risks at the study site. The detected land-use transformations unveiled different historical moments, culminating in a landslide disaster (804 houses destroyed, 3000 families at risk). This catastrophe necessitated a radical and significant intervention, incurring substantial costs for the megacity administration (USD 26.05 million). This landslide was the largest recorded in the megacity and one of the most extensive in urban areas across Latin America.

Application of Artificial Intelligence and Sensor Fusion for Soil Organic Matter Prediction.

Soil organic matter (SOM) is one of the best indicators to assess soil health and understand soil productivity and fertility. Therefore, measuring SOM content is a fundamental practice in soil science and agricultural research. The traditional approach (oven-dry) of measuring SOM is a costly, arduous, and time-consuming process. However, the integration of cutting-edge technology can significantly aid in the prediction of SOM, presenting a promising alternative to traditional methods. In this study, we tested the hypothesis that an accurate estimate of SOM might be obtained by combining the ground-based sensor-captured soil parameters and soil analysis data along with drone images of the farm. The data are gathered using three different methods: ground-based sensors detect soil parameters such as temperature, pH, humidity, nitrogen, phosphorous, and potassium of the soil; aerial photos taken by UAVs display the vegetative index (NDVI); and the Haney test of soil analysis reports measured in a lab from collected samples. Our datasets combined the soil parameters collected using ground-based sensors, soil analysis reports, and NDVI content of farms to perform the data analysis to predict SOM using different machine learning algorithms. We incorporated regression and ANOVA for analyzing the dataset and explored seven different machine learning algorithms, such as linear regression, Ridge regression, Lasso regression, random forest regression, Elastic Net regression, support vector machine, and Stochastic Gradient Descent regression to predict the soil organic matter content using other parameters as predictors.

Using Hyperspectral Images and Lidar Data to Create Models for the Classification and CAVE Visualization of Tree Species

The article describes the use of artificial intelligence methods to create models for detecting and classifying tree species, using hyperspectral images and LiDAR data in the aerial photography of energy line structures. The most important output is a validated model with cloud infrastructure support for detecting and classifying objects of interest at the TRL 5 level, which is also exceptional on a global scale. The outputs of the research are also a geodatabase of reference tree characteristics, a library of spectral curves, a database of simulation of tree growth, but also a cloud infrastructure to support the development of classification models and data storage. An important output will be the visualization of the results of simulations in the "CAVE" environment.

Deformation Patterns and Failure Mechanisms of Soft-Hard-Interbedded Anti-Inclined Layered Rock Slope in Wolong Open-Pit Coal Mine

Since the beginning of spring 2022, successive landslides have occurred in the eastern pit slope of the Wolong Coal Mine in Qipanjing Town, Otog Banner, Inner Mongolia, which has adversely affected the mine’s production safety. This study aims to reveal the deformation patterns and failure mechanisms of landslides. Firstly, this study establishes the stratigraphic structure of the eastern pit slope of the Wolong Coal Mine using extensive field geological surveys combined with unmanned aerial vehicle photography, drilling, and comprehensive physical exploration techniques. Indoor geotechnical tests and microscopic experiments reveal that rock mass typically exhibits the characteristics of expansibility and water sensitivity. Moreover, the mechanical parameters of the rock mass were determined using a combination of the window sampling method, the Geological Strength Index, and the Hoek–Brown strength criterion estimation theory. Finally, this study consolidates the previously mentioned insights and employs FLAC3D (7.0) software to assess the stress–strain characteristics of the excavated slope. The results indicate that the deformation mode of the Wolong open pit coal mine is the toppling failure of soft-hard-interbedded anti-inclined layered rock slopes. The unloading effect and rock expansion-induced softening lead to stress concentration at the slope corners and more substantial deformation, thereby accelerating upper slope deformation. The deformation and destabilization process of landslides is categorized into four stages: the initial deformation stage, the development stage of lateral shear misalignment, the development stage of horizontal tensile-shear damage, and the slip surface development to the preslip stage. This research offers valuable references and engineering insights for future scientific investigations and the prevention of similar slope-related geological hazards.

THE IMPACT OF MINING EXCAVATIONS (VOIDS) ON THE WATER QUALITY MANAGEMENT IN THE NORTH KALIMANTAN PROVINCE

The presence of a coal mining business permit not only has an economic impact but can also cause environmental problems. The objectives of this research are (1) to analyze the impact of PT XXX's overflowing voids on water management and water quality, and (2) to carry out spatial analysis, mapping and estimation of coal reserves that cannot be extracted by PT YYY. The research method is a case study selected by purposive sampling at two companies in North Kalimantan Province using water quality test analysis with grab sampling and estimating the potential loss of coal reserves using the rule of gradual changes and rule of nearest point methods as well as aerial photography. The research results show that in general the water use meets Class 3, namely: Water that can be used for cultivating freshwater fish, animal husbandry, water for irrigating crops, and/or other uses that require the same water quality as that use. Apart from that, PT XXX has built an overflow water channel into the PT YYY area without management approval, thus having a negative impact on PT YYY's mining plans which are dominated by peat land. Map of PT YYY's lost coal potential of 714,688 MT with an affected area of 15.16 hectares with details of 6.4 ha being the void area included in PT YYY's IUP area, 1.86 ha being the road and buffer zone area and 6.9 ha is a potential lost area of PT YYY.

Thickness Estimation of the Soil‐Sedimentary Cover Inside Causewayed Enclosures to Locate an Occupation Layer: Map of the Archaeological Potential of the Neolithic Causewayed Enclosure of Le Pontet

ABSTRACTThe western centre of France is one of the richest regions of Western Europe in terms of causewayed enclosures from the Neolithic period. To date, more than 300 such sites have been identified. Most causewayed enclosures in the region range in size from some hectares to over 10 ha. Exhaustive excavation of such sites is rarely performed because such operations are both financially expensive and time intensive. Completing an exhaustive excavation of these sites is also extremely complicated due to their complexity. Recording the most complete possible site plan is a major challenge for optimizing excavation. Traditionally, aerial photography has been the primary method used for delineating causewayed enclosure plans. A magnetic survey can also provide complementary information. Magnetic imaging reveals both enclosure ditches and internal features (pits, postholes, etc.) quickly and with high spatial resolution. At some sites, occupation layers dating from the time of enclosure may be preserved and contain archaeological artefacts or small features in situ. This article proposes a protocol for locating a Neolithic occupation layer inside a Neolithic causewayed enclosure. To locate the areas where this layer is likely to be present, a map of the archaeological potential of the Le Pontet site was produced based on a thickness map of the soil‐sedimentary cover. This map was created by combining an apparent electrical resistivity map, the results of electrical resistivity tomography, an orthophotograph with contrasting cropmarks and the results of dynamic cone penetration tests. To validate the archaeological potential map, an excavation campaign was conducted in 2020 to investigate several sectors; the aim was to prove the presence of the occupation layer and study the pedo‐sedimentary stratigraphy of the site.

Enhancing water balance assessment in urban areas through high-resolution land cover mapping: Case study of Debrecen, Hungary

Land cover (LC) mapping in urban areas is one of the core applications in remote sensing (RS), and it plays an important role in modern urban planning and management. In order to support up-to-date simplified water balance calculation, LC modelling needs to be developed for precise and high-resolution operations to monitor rapidly expanding urban areas. This study provides a spatial decision support tool for urban water balance calculations considering hydrological parameters at pixel scale. The study is aimed to create a high-resolution LC map for more precise water balance calculations to identify and link the most important parameters, such as crop coefficient, for estimating crop evapotranspiration, runoff coefficients, and infiltration rate. Meteorological data from 2016 to 2019 were involved in evapotranspiration estimation. The study site is Debrecen, a city in northeast Hungary with a population of about 200,000. By integrating Landsat 8 imagery, Google Earth (GE), and Corine Land Cover (CLC), a LC map of 30 m resolution was created with 81.2 % overall accuracy (OA) and a coefficient of kappa greater than 0.78. For the classification of Landsat 8, seven classes were assigned (forests, sealed surfaces, areas with crop cover, grassland, semi-sealed surfaces, bare ground and surface water bodies). Classification results were validated by 101 ground truth samples using other satellite images and aerial photographs. Water balance parameters (i.e. surface runoff, infiltration and evapotranspiration) were then defined and calculated at a pixel scale and for larger areas. The technique developed in this study can also be utilized in other urban areas.

Intensively Cultivated Roman Villae Estates: Case Study of Medulin Bay (Istria, Croatia)

ABSTRACTOur image of the Roman landscape of Istria is characterised by large‐scale centuriation and architectural remains of Roman villae. Detection and mapping of other, less dramatic landscape features require systematic large‐scale prospection, but this faces significant difficulties in the Mediterranean environment. However, the developments in the field of airborne laser scanning offer the possibility to create archaeologically usable digital terrain models under water and under very dense and low maquis vegetation. This paper reports on the use of terrain models created using a green laser and a sophisticated archaeologically driven ground point filtering strategy. Combined with archaeological aerial photo interpretation, this provides the means for landscape mapping and interpretation that has revealed a wealth of archaeological structures hinting at Roman agricultural practices and landscape. Our case study is based on a laser scan of about 24 km2 of land and underwater terrain in Medulin Bay. Processed, visualised and interpreted for archaeological purposes, the data reveal not only features ranging from prehistoric hilltop settlements to modern military installations but also features a complex picture of the Roman land use. Of particular interest is the large number of planting pits, which extend over a total length of 4 km. They were laid on a regular grid of approximately 35 × 35 m, sometimes combined in contiguous parcels. They can be interpreted as remains of orchards or tree nurseries of Roman date, and the paper examines also the question of whether they can be linked to the associated Roman estates. The case study area presented here shows that the potential of remote sensing methods goes far beyond the mere finding of traditional sites but can open up new landscape‐scale perspectives on regions that have been archaeologically little explored.

Long-Term Volumetric Change Estimation of Red Ash Quarry Sites in the Afro-Alpine Ecosystem of Bale Mountains National Park in Ethiopia

The Bale Mountains National Park (BMNP) in Ethiopia comprises the largest fraction of the Afro-Alpine ecosystem in Africa, which provides vital mountain ecosystem services at local, regional, and global levels. However, the BMNP has been severely threatened by natural and anthropogenic disturbances in recent decades. In particular, landscape alteration due to human activities such as red ash quarrying has become a common practice in the BMNP, which poses a major environmental challenge by severely degrading the Afro-Alpine ecosystem. This study aims to quantify the long-term volumetric changes of two red ash quarry sites in the BMNP using historical aerial photographs and in situ data, and to assess their impact on the Afro-Alpine ecosystem. The Structure-from-Motion multi-view stereo photogrammetry algorithm was used to reconstruct the three-dimensional landscape for the year 1967 and 1984 while spatial interpolation techniques were applied to generate the current digital elevation models for 2023. To quantify the volumetric changes and landscape alteration of the quarry sites, differences in digital elevation models were computed. The result showed that the volume of resources extracted from the BMNP quarry sites increased significantly over the study period from 1984 to 2023 compared with the period from 1967 to 1984. In general, between 1967 and 2023, the total net surface volume of the quarry sites decreased by 503,721 ± 27,970 m3 and 368,523 ± 30,003 m3, respectively. The extent of the excavated area increased by 53,147 m2 and 45,297 m2 for Site 1 and 2, respectively. In terms of habitat loss, major gravel road construction inside the BMNP resulted in the reduction of Afro-Alpine vegetation by 476,860 m2, ericaceous vegetation by 403,806 m2 and Afromontane forest by 493,222 m2 with associated decline in species diversity and density. The excavation and gravel road construction have contributed to the degradation of the Afro-Alpine ecosystem, especially the endemic Lobelia rhynchopetalum on the quarry sites and roads. If excavation continues at the same rate as in the last half century, it can threaten the whole mountain ecosystem of the National Park and beyond, highlighting the importance of preventing these anthropogenic changes and conserving the remaining Afro-Alpine ecosystem.

Preference aggregation method in determining brightness threshold values for object recognition on optical images

Relevance. The need to automate the processing large volumes of geophysical information extracted from images obtained through aerial photography or space remote sensing. The main stage of image processing for object recognition is segmentation. It consists in dividing the image into two homogeneous, clearly distinguishable non-overlapping areas with known boundaries. Threshold segmentation methods are simple but effective and therefore popular. Aim. To describe the experimental studies results of application of the robust method of interval fusion with preference aggregation previously developed by the authors for selecting threshold values when segmenting objects in optical images. Objects. Coastline images taken by Earth remote sensing satellites. Methods. The analyzed image is divided into equal bands. Based on the brightness histogram of each band, its characteristic brightness interval is determined. For the obtained brightness intervals, the fusion result is calculated using the interval fusion with preference aggregation method. The latter is used as a threshold brightness value when forming a segmented image. Results. The results of experimental studies of 100 satellite images of the coastline showed that the proposed method provided correct separation of land and sea regions in 84 images. And the traditional methods of maximum entropy, arithmetic averaging and Otsu showed correct results only in 27, 76 and 73 cases, respectively. Evaluations of such the metrics as Precision, Recall and deviation of the calculated threshold from a known correct value showed the dominance of the proposed method over other tested ones in terms of segmentation quality.

Image Filtering to Improve Maize Tassel Detection Accuracy Using Machine Learning Algorithms.

Unmanned aerial vehicle (UAV)-based imagery has become widely used to collect time-series agronomic data, which are then incorporated into plant breeding programs to enhance crop improvements. To make efficient analysis possible, in this study, by leveraging an aerial photography dataset for a field trial of 233 different inbred lines from the maize diversity panel, we developed machine learning methods for obtaining automated tassel counts at the plot level. We employed both an object-based counting-by-detection (CBD) approach and a density-based counting-by-regression (CBR) approach. Using an image segmentation method that removes most of the pixels not associated with the plant tassels, the results showed a dramatic improvement in the accuracy of object-based (CBD) detection, with the cross-validation prediction accuracy (r2) peaking at 0.7033 on a detector trained with images with a filter threshold of 90. The CBR approach showed the greatest accuracy when using unfiltered images, with a mean absolute error (MAE) of 7.99. However, when using bootstrapping, images filtered at a threshold of 90 showed a slightly better MAE (8.65) than the unfiltered images (8.90). These methods will allow for accurate estimates of flowering-related traits and help to make breeding decisions for crop improvement.