Proceedings of UASG 2021: Wings 4 Sustainability

The availability of Very High-Resolution (VHR) remote sensing datasets from the Unmanned Aerial Vehicle (UAV) based sensors are changing the methods of cartographic mapping as well as visualization by taking advantage of both the high spatial resolution as well as high radiometric resolutions. A high-fidelity digital elevation model (DEM) can be prepared using these UAV datasets, which can produce high-quality orthoimages. In the present study, the space-borne lidar elevation datasets from the Ice, Clouds, and Land Elevation Satellite (ICESat-1) and TanDEM-X 90 m DEM from TerraSAR-X add-on for Digital Elevation Measurement (TanDEM-X) mission are utilized for the comparison of elevation values from DEM generated using UAV datasets for the experimental site in Switzerland. The experimental site is part of Yverdon-Les-Bains, which is a municipality in the district of Jura-Nord VauDOIs, canton of Vaud, Switzerland. The openly accessible dataset from the Sensefly Sensor Optimized for Drone Applications (S.O.D.A.) includes 235 true-color RGB images acquired from a flight height of 106 m, at an average Ground Sampling Distance (GSD) of 2.64 cm. The datasets are processed in Pix4D software for the bundle block adjustment, followed by the generation of DEM and orthomosaic. The comparison of ICESat-1 elevation data with DEM depicts a difference of about 26 cm on plain ground, which is reasonably good considering the use of a Global Navigation Satellite System (GNSS) network in Real-Time KinematicReal-Time Kinematic (RTK) mode. The quality report depicts the mean of geolocation accuracy in X, Y, and z as 2.73 cm, 2.73 cm, and 3.46 cm respectively, which is practically highly accurate. Root Mean Square Error (RMSE) in X, Y, and z is computed as 1.7 cm, 2.27 cm, and 2.31 cm respectively. The study depicts that practically the cartographic potential for the UAV dataset is suitable for mapping at a scale range of 1:250 to 1:300 or better for such plain terrain conditions, meeting the engineering drawing requirements for facility management and utility mapping.

The precise boundaries of the cadastral parcels from the Unmanned Aerial Vehicle (UAV) data are essential for any eGovernance application. The pix2pix, image-to-image translation using the conditional Generative Adversarial Network (cGAN) models, has emerged as an alternative to the traditional machine learning and image processing algorithms. It has been used and demonstrated for productive purposes in different domains without any change in the pix2pix network model and loss functions. The pix2pix model is implemented in this research for extracting the cadastral parcel boundaries using the existing UAV data set, and the corresponding digitised data. The input data set is prepared using the python modules. The model is also used to predict the synthetic UAV data from the map data. The predicted boundary of the model is very useful. The proposed model can reduce the manual labour and human interventions in outlining the parcel boundary from UAV data.

Expanding needs for using UAV (Uncrewed Aerial Vehicles) remote sensing approaches, such as terrain-following aerial mapping applications using LiDAR (light detection and ranging) in catastrophic applications. New extracts in UAV mapping still contain a limited number of studies for analyzing fine-scale mapping accuracy in UAV remote sensing methods—terrain-following aerial mapping for UAVs based on external airborne LiDAR integrated with the flight controller. We introduce the UAV system for the terrain after mapping the high-rise area by circumventing obstacles around it, expanding it so that UAVs flying at low altitudes can collect high-quality ground information while protecting them from all kinds of obstacles up and down. A more informative map is prepared to speed up the rescue and relief operations of the devastated area. Then on-destructive techniques have been surveyed and explored Solani riverbank sites of the high undulating area in Roorkee, Haridwar district in the Indian state of Uttrakhand. This work aims to critically analyze fine-scale remotely sensed data for mapping using LiDAR and UAV obtained the structure from motion photogrammetry. This work outlines the approaches applied to remote sensing data to reveal potential sensitivities, reflecting the close visual methods of the catastrophic region.

In this work, a Convolutional Neural Network based approach is presented for accurate classification of forest areas with fire from UAV images. In general, the deeper the CNN architecture, the classification of ‘fire’ versus ‘no fire’ is more accurate. However, deeper architectures consume lot of battery power and impose constraints on the processor used in UAV. It is time taking too. Hence, architectures like ResNet50 are not suitable as 23 million parameters are required to train a ResNet50 model. In this regard, mobile CNN architectures are quite handy and they require very few parameters of typical 1–7 millions. They are faster also and take very less time for inference. In this work, the features from selected pre-trained mobile CNN architectures i.e., Squeezenet, MobileNetv1, MobileNetv2, MnasNet, MobileNet v3, SqueezeNext, ShuffleNet, CondenseNet, DiCENet, FBNet, MixNet, and EfficientNet Lite-0, EfficientNet Lite-1 are used in the classification process. All the architectures are pre-trained on ‘imagenet’ dataset with 1000 classes and 14 millions of images. Features from the last pooling layer of each network are obtained. Feature fusion (concatenation) from the selected mobile CNN architectures is considered for classifying the images with ‘fire’ and ‘no fire’. SVM classifier is applied to the fused feature vector. In general, as the size of the fused feature vector increases, the classification accuracy increases. A wildfire image dataset with 2096 images is chosen with balanced classes of ‘fire’ and ‘no fire’. With a 80% train and test split, the mean classification accuracy obtained is in excess of 98%. Various other performance metrics are also given to emphasize the merit of the proposed approach.

Automatically extracting buildings from remotely sensed imagery has always been a challenging task, given the spectral homogeneity of buildings with non-building features as well as the complex structural diversity within the image. Traditional machine learning (ML) based methods deeply rely on a huge number of samples and are best suited for medium-resolution images. Unmanned aerial vehicle (UAV) imagery offers the distinct advantage of very high spatial resolution, which is helpful in improving building extraction by characterizing patterns and structures. However, with increased finer details, the number of images also increases many folds in a UAV dataset, which require robust processing algorithms. Deep learning algorithms, specifically Fully Convolutional Networks (FCNs) have greatly improved the results of building extraction from such high resolution remotely sensed imagery, as compared to traditional methods. This study proposes a deep learning-based segmentation approach to extract buildings by transferring the learning of a deep Residual Network (ResNet) to the segmentation-based FCN U-Net. This combined dense architecture of ResNet and U-Net (Res-U-Net) is trained and tested for building extraction on the open source Inria Aerial Image Labelling (IAIL) dataset. This dataset contains 360 orthorectified images with a tile size of 1500 m2 each, at 30 cm spatial resolution with red, green and blue bands; while covering total area of 805 km2 in select US and Austrian cities. Quantitative assessments show that the proposed methodology outperforms the current deep learning-based building extraction methods. When compared with a singular U-Net model for building extraction for the IAIL dataset, the proposed Res-U-Net model improves the overall accuracy from 92.85% to 96.5%, the mean F1-score from 0.83 to 0.88 and the mean IoU metric from 0.71 to 0.80. Results show that such a combination of two deep learning architectures greatly improves the building extraction accuracy as compared to a singular architecture.

People's failure to maintain a social distance is causing the COVID19 virus to spread. We have used the drone thermal images for a maximum of 10 km of coverage to detect temperature and reduce virus spread areas. The part of the work is based on utilizing disinfectant spraying drones, disinfectant testing with the guidance of doctors, setting the path planning of drones for surveying the temperature of people, and monitoring the infected place using GPS. When the thermal camera of the drone detects the temperature values using remote sensing images, the drone covers crowded places like hospitals, cinemas, and temples using remote sensing images. One drone model is designed to provide present results using thermal images. The Proposed drone can cover an affected area of up to 16,000 square meters per hour for capturing remote sensing images. It predicts affected areas using faster CNN algorithms with 2100 thermal images. Thermal mapping is used to monitor the social distance between people, alert people that a virus is spreading, and reduce the risk factor of people's movement. In this paper, remote sensing images are analysed and detect higher temperature areas using thermal mapping (Messina and Modica in Remote Sensing 12:1491, 2020).

As the rapid development is being focused in the urban area, there is a need for the utilisation of a system for updating this profile immediately. The usage of Uncrewed Aerial Vehicle (UAV) for mapping purposes is one of the current technologies being used in recent years. UAVs are widely used in a variety of domains due to their low price, ability to deliver very high resolution data, and ability to fly at low altitudes without being constrained by overcast weather. Typically, data extraction methods for UAVs are still quite limited, and traditional approaches are still used. For mapping applications, orthoimage features are often manually recognised and digitised using visual interpretation skills. Unfortunately, these approaches are time-consuming, costly, and repetitive. Pixel-based classification approach is frequently utilised to help extract low-level features, in which the image is categorised only based on spectral characteristics. The drawback of this approach is that the pixels in the overlapping region are misclassified as a result of class confusion. Moreover, pixel based classification performs very poorly in high resolution images. The Object-Based Image Analysis (OBIA) classification technique has large potential for automatic data extraction from Very High Resolution (VHR) images. OBIA techniques start with segmentation of image followed by classification and feature extraction using contextual information and rule base. In this study, an attempt is made to assess the capability of OBIA for detailed classification of highly dense urban areas mapped by UAV with a VHR imagery of the order better than 5 cm. The image is segmented using multiresolution image segmentation with a suitable scale, compactness and smoothness to form homogeneous image objects. Various parameters (spectral, texture, context and elevation) are computed for the VHR UAV Images. Rules are formulated to extract and categorise urban features specifically for roads and buildings. The segmented roads are classified into categories based on width and connectivity. Buildings extracted are categorised based on their elevation and size. The study efficiently demonstrates the potential of VHR orthoimage and Digital Surface Model (DSM) for urban classification using the OBIA techniques. The finest of details captured by UAV can be effectively classified using the segmentation and classification approach.

A massive wave of urbanisation has grasped both developed and developing nations in the last decade. Various studies showcase the rising trend of rapid urbanisation growth due to demographic shifts, catalysed by numerous global and local parameters. India is professed to house 50 percent of its population in urban areas by 2030. The urban citizen’s rising expectations regarding infrastructure, amenities and safety will dovetail with this phenomenon. To cater to the demands of increased urbanisation, the concept of Smart cities is being advocated as an apt solution. Even so, there is widespread ambiguity about the challenges associated with adopting smart solutions into the existing urban social form. Though there have been many tentative explanations regarding the basic framework of a smart city, there are no known validated definitions. The main notable feature is the use of information and communication technologies (ICTs) within the cyclic dynamics of a city to facilitate the smooth and cost-effective working of urban areas. This ICT infrastructure enables real-time data collection, analysis, response, and storage. This has been deemed beneficial due to the increase in efficiency of operation and management services involved with the day-to-day functioning of any urban area. Smart cities advocate using information and communication technologies (ICTs) within the dynamics of a city to enable the real-time collection, analysis, and storage of big data. This is beneficial due to the increased efficiency of operation and management services involved with an urban area’s daily functioning. One such technological intervention is the ‘drone’ or the ‘unmanned aerial vehicles (UAVs)’. UAVs have a wide variety of uses in a smart urban fabric, from geospatial integration to traffic management, surveillance, disaster response, etc. In the current nascent stage of research and development of UAVs as one of the innovative solutions for smart cities in India, questions arise regarding privacy, cost of production, technical knowledge, safety and security with their large-scale use. This paper aims to assess the applicability of UAVs in overall smart city planning and management. The feasibility analysis method is adopted to analyse the felicitousness of UAVs in a smart city’s planning and design phases. The results of the study undertaken in this paper highlight the challenges and opportunities in the planning and management of smart cities by integrating UAVs. The paper enumerates the relevance and appropriate benefits of using UAVs to plan, design, and perpetuate Indian smart cities.

Whether planning urban development, conducting a hydrological construction in different terrain conditions, or analyzing terrain features for oil and gas exploration: accurate elevation information is vital. Digital Terrain Model Digital Terrain Model Digital Terrain Model(DTM) and Digital Surface Model (DSM) is the elevation model that provides elevation information of terrain and earth features (object), respectively. This study aims to assess the elevation accuracy of some of the most preferred Uncrewed Aerial Vehicle (UAV) data processing software from ESRI, Pix4DsoftwarePix4Dmapper, and Agisoft Photoscan. In this study, DJI Phantom 4 is used to collect 147 very high-resolution overlapped images in the selected study area (Department of Civil Engineering, IIT-Roorkee, India) at the defined height. Collected images are processed using all the selected platform elevation datasets, i.e. Digital Surface Model Digital Surface Model DSM(DSM) is generated. Vertical elevation error is estimated by elevation profile and statistical comparisons of UAV-derived elevation in the DSM datasets. This study helps to select the best UAV data processing software for the project that requires high elevation accuracy in topographical mapping or urban object utilization.

Advances in UAV technology and processing made it feasible to obtain ultra high-resolution (UHR) imagery and three-dimensional (3D) data, which can be efficiently used for urban forestryurbanurban forestry, green space mapping, green infrastructure planning and monitoring with high accuracy and at a cost-effective manner. The conventional feature extraction image analysis techniques fail on UHR UAV Images as the geometry of features is very well defined and characterized by a very heterogeneous texture. The approach required for such ultra-high-resolution images should support the cognitive analysis that we use in visual image interpretation, which is a form of knowledge-driven analysis incorporating shape, texture, pattern and contextual information. The present study aims at delineating and extract vegetation types and height estimation using UHR UAV images in parts of urban slums and dense urban environments. The methodology involved utilizes a multiresolution image segmentation to create basic image objects at a scale that allows for homogenous object extraction while maintaining variability. Normalized Differential GreenNormalized Differential Green Vegetation Vegetation (NDGVNDGVNormalized Differential Green Vegetation) and Visible-Band Difference Vegetation IndexVisible-Band Difference Vegetation Index (VDVIVDVIVisible-Band Difference Vegetation Index) band ratios are computed by combining the red, green and blue (RGB) spectrum of the UHR image. Digital Surface Model (DSM) and Digital Terrain Model (DTM) were input to the canopy height model (CHM) for height estimation and refinement of different vegetation categories. Crown shape parameters and texture parameters are tested and compared for vegetation characterization and categorization. Results show the 3D data set derived from UAV UHR imagery’s potential in detection of treetops, delineation of vegetation and its categorization.

This study aims to present an environmental gaseous sensing analysis using drones in urban development control for the industrial area. The data collection method is based on the possibility of gas dispersion in a heavy industrial area in Klang, Malaysia, to the neighbouring land uses. The sniffer has carried five types of gaseous sensors and is mounted in DJI Matrice 100 quadcopter UAV. However, for this study, we analysed two significant gases related to an industrial area consisting Carbon dioxide (CO2) and Hydrocarbon (CxHy). The information has been collected in two modes of time which are in the early morning and afternoon. The data were mapped and analysed with a vector layer to identify whether it breaches concentration limits for gases collected. The finding stated that the morning concentration reading is denser compared to the afternoon. Results show that CO2 and CxHy are still under control and minimise the risk for the local population. However, the safety precaution should be undertaken since gas dispersion's future potential would go beyond and affect the surrounding activities. In conclusion, this study shows the UAV’s potential as one of the best mechanisms to monitor the environmental effect. Simultaneously, there is a need to review existing urban development control since climate change and sustainability are linked through their interaction in industries, and their surrounding land uses.

A clean Environment is the basic right of every human being. For the last two decades, India is dealing with huge environmental issues and one of the major issues is waste management. There is a huge gap between the generation and processing of waste in India. Some of the current methods of dealing with waste disposal are causing more harm than benefit to the environment and public health. Huge advancement in technologies such as Artificial Intelligence, Machine learning, smart sensors, IoT, UAVs or Drones, automation, etc. has improved human life in every domain such as healthcare, Agriculture, Consumer Services, and Manufacturing. There is huge scope in harnessing the potential of these technologies in the field of Waste Management. Due to the advancement in sensors and availability of cost-effective commercial Drones, Drones have shown significant improvement and huge application scope in different domains. There is a huge prospect of using drones in waste management. It is one of the popular tools and technology, whose use can be explored in different areas of Waste Management. Government and Waste Management organizations can utilize smart UAVs or Drones to efficiently manage waste disposal at Landfills and dumping zones. In this paper, an attempt has been made to review and identify the different areas of waste management where Drones can be utilized efficiently in India. This article also reviews the various challenges and requirements of using Drone technology in waste management.

Many buildings are using solar panels as an additional source of electricity. As solar energy is renewable energy and the maintenance cost of solar panels is cheap. This research uses a statistical approach of analyzing point clouds generated from UAV-based photogrammetric processing. An algorithm has been developed to extract solar panels on the building rooftops. The data acquisition is done using an Unmanned Aerial Vehicle (UAV) platform mounted with an optical sensor. The RGB images acquired are further used to generate a photogrammetric point cloud dataset. Geomatics engineering building of Indian Institute of Technology Roorkee, India is considered as the study area, on which solar panels were already installed on its roof. Normal vectors are computed for each points in the building point cloud dataset. The normal vector has its components in the x-axis, y-axis, and z-axis correspondingly. Based on the contribution of the z-component of normal vectors, the points are classified into roof, facade, and solar panel points respectively. The results obtained are evaluated by comparing classified points with respect to manually classified solar panel points. This comparision suggests that the developed algorithm is effective in extracting the solar roof panels efficiently. This research can be used to calculate the effective area of solar panels.

Crime is heterogeneously distributed and occurs at the most vulnerable places. Crime occurs under poor surveillance and safety, due to lack of public protection and results in damage to public property or human life, and creates a public discrepancy in that particular location. Crime is disastrous because of its unpredictability and unpreparedness for enforcement officers. Finding the probability of occurrence of crimes within such vulnerabilities will help us to deploy certain countermeasures to reduce crime. Crime is limited to location and place. Geographically crime can be considered as a function of lack of surveillance, delay in mobility and control, and probably hidden escape paths utilized by criminals. In this research, a Spatio-Statistical Model was developed for probability-based Crime Prediction using past data and location intelligence technology. Neighborhood Analysis was performed to evaluate the clustering distance between individual crime occurrences within Vadodara city and individual police stations in the neighborhood. The spatial distance is converted into Geographical Coordinate System to calculate latitudinal and longitudinal extents of crime zones in each taluka of the city, which is then utilized to create the Interpolated probability raster for each crime zone with a pixel value equivalent to the probability of occurrence of crime in that location. The Inverse distance weighted (IDW) interpolation technique generated an interpolated surface which was then represented spatially with quantile divisions to form probability zones with the adjoining nearest police jurisdiction. This will enable law enforcement officers to make probability-based surveillance decisions while incorporating the past data intelligence, time of occurrence of crime, and make efficient serviceable patrolling routes and improve crime control with minimal resources. Using this model, the police officers will be able to create patrol routes based on time and zone of highest probability of crime, to ensure safety. The time-based probability of crime is also calculated using the Bayesian probability formula to get the peak crime hours so that surveillance need to be increased at the appropriate time. UAVs mounted with thermal vision can be deployed in the generated high probability zones at the highest probable time of the crime, to monitor the situation aerially without alarming the criminals. In this research it is created an open-source pixel-based route selection algorithm that could identify hotspot locations of crime so that law enforcement officers can watch human movements and follow them silently using UAV’s thermal camera in nighttime also to obtain their hideouts and catch criminals.

Remote Sensing data, either airborne or satellites, are very much useful for incorporating the Geographical Information System (GIS) technology. SAR sensors are good as compared to optical sensors for monitoring maritime activity due to their capability of penetrating clouds and can work without depending upon any weather condition. SAR sensors can work day and night while optical sensors need a source to illuminate the surface hence can only work in the daytime. Many studies have been done on UAV SAR sensors for different applications like oil spills, ship detection, etc. Moreover, the polarimetric technique helps in understanding the feature much more in detail by using phase information like orientation and shape of the object using scattering behavior. In this paper, the main focus of the study is the Automatic ship detection using the Adaptive Threshold Algorithm popularly known as Constant False Alarm Rate (CFAR) for polarimetric UAV SAR data. Coherency Matrix ( $${T}_{3}$$ ) is computed from quad-pol covariance SAR data $${C}_{3}$$ and CFAR algorithm is applied to each element of the coherency matrix to detect ships. The sea surface follows the surface scattering and this can be highly helpful to distinguish the ships from the sea background. Moreover, due to the homogeneous background of imagery, the CFAR algorithm works more precisely as it can compute the adaptive threshold for each pixel using the background area by assuming it to the Gaussian in nature. Moreover, the Global Self-consistent, Hierarchical, High-resolution Geography Database (GSHHG) vector coastline layer and Digital Elevation Model (DEM) are used for masking out the land area to enhance the area of interest. In this study, $${T}_{22}$$ element of the scattering matrix shows better results in the detection of the ships and in determining the shape of the ships. Finally, the efficiency of the algorithm is measured using the Receiver Operating Characteristics (ROC) curve.

Advancements in unmanned aerial vehicle (UAV) technology have enabled the acquisition of images of a geographical area with higher spatial resolutions as compared to images acquired by satellites. Detection and segmentation of objects in such ultrahigh-spatial-resolution (UHSR) images possess the potential to effectively facilitate several applications of remote sensing such as airport surveillance, urban studies, road traffic monitoring crop monitoring, etc. Investigating these images for target extraction tasks turns out to be quite challenging, in the terms of the involved computation complexities, owing to their high spatial resolutions and information content. Due to the development of several deep learning algorithms and advanced computing tools, there exists a possibility of harnessing this information for computer vision tasks. Manual surveillance of airports or similar areas and manual annotation of images are cost-intensive and prone to human-induced errors. Therefore, there exists a substantial requirement of automating the task of keeping track of the airplanes parked on the premises of airports for civil and military services. With this paper, we propose a framework for detecting and segmenting such airplanes in UHSR images with supervised machine learning algorithms. To detect the target i.e., airplanes, MobileNets-deep neural network is trained, whereas to segment the target, U-Net-convolutional neural network is trained with our dataset. Further, the performance analysis of the trained deep neural networks is presented. The UHSR image dataset utilized in this research work is an airport dataset provided by SenseFly. Data is acquired by eBee classic drones, flying at a height of 393.7 ft., which provide 2D-RGB images with a ground resolution of 3.14 cm/px.

The dynamic development of 3D building reconstruction using digital images obtained with unmanned aerial systems (UAS) has been observed in recent years. The popularity of UAS is due to its wide technological availability at a low price, compared to geodetic measurement equipment, laser scanners, or manned flight missions. In practice, the usage of UAS for 3D building reconstruction and modelling allows the acceleration of the production process (image acquisition, processing, computation) while maintaining a high quality of the final product. With the increasing number of new flying objects in airspace and because of differences in UAS regulation in each EU country, it was necessary to adapt the rules for the operation of unmanned aircraft to standardize regulations, make operations easier, and assure aviation safety. Due to this fact, from 31 December 2020, the new European Union (EU) Commission Implementing Regulation 2019/947 on the rules and procedures for the operation of unmanned aircraft entered into force across the continent. The new regulations replaced each EU national’s existing laws and applied to all UAS pilots. They have adopted a risk-based approach and do not distinguish between leisure or commercial activities like previous regulations. To assess the operational risk and to determine the category of flight mission, the weight and specifications of the UAS, the operation, and UAS pilot qualification are taken into account. Because of that, new categories of operations have been established. In this study, the review of 3D reconstruction using UAS was performed and the new EU UAS regulations in the context of the image acquisition of buildings in different levels of detail (LoD) were studied. For this purpose, the practical 3D reconstructions of buildings were analyzed. Furthermore, taking Poland as an example, new unified EU rules were compared with the previous ones.

The continued understanding of the influence of flight planning characteristics on data quality is crucial in the demand for minimizing costs and maximizing the output potential of Uncrewed Aerial Systems (UAS) for forestry applications. This study was conducted to ascertain the effects of various combinations of flying height and percentage overlaps on the quality of photogrammetry data products generated from images acquired by a low-cost UAS (Mavic 2 Pro), with emphasis on tree crown delineation in a Mangium plantation forest in the Philippines. The quality of the products is evaluated based on their completeness and the accuracy of tree crown delineations. Results suggest that the percentage completeness increases as the flying height and percentage overlap increase. More than 90% completeness was achieved for 90% overlap regardless of the flying height. Tree crown delineations using multiresolution segmentation of Digital Surface Models (DSMs) generated from images with a flying height of 120 m and percentage overlap of 80% and 90%, achieved the highest overall accuracy of 43.35%. This study showed that a minimum of 80% overlap must be aimed when acquiring images to ensure higher completeness of the data products and that flying at 120 m above ground with at least 80% overlaps can provide more accurate tree crown delineations.

The 3D Point cloud derived especially from drone-derived images is highly unstructured, redundant and has varied density. These point clouds need to be segmented and classified into different groups representing similar characteristics in the scene presented which is a challenging task especially when the 3D scene contains a mix of varied man-made or unstructured natural scenes such as vegetation etc. Successful operation of such technology will lead to a wide variety of remote sensing, computer vision and robotics applications. In this paper, we have used a hybrid approach for effective segmentation of the point cloud. The combination of RANSAC, DBSCAN and Euclidean method of Cluster Extraction proved to be useful for precise segmentation and classification of the point cloud.

Digital Terrain Models (DTMs), which represent the topography of the bare Earth surface, are widely used in many geomatics applications. In parallel to the emergence of sophisticated Unmanned Aerial Systems (UASs) in recent years, they are produced from point clouds generated through aerial images taken from digital imaging systems mounted on UASs. The first and most important step of DTM production is to remove the points of the above-ground objects such as trees, buildings, bridges, etc. A great variety of point cloud filtering strategies have been developed so far. However, due to the irregularities in the topography of the Earth's surface, all proposed approaches employ several user-defined parameters, which makes point cloud filtering dependent on the parameter values defined. Since complex topographies make it very hard to define some protocols to estimate the best parameter values, users usually have to try a large number of parameter values for optimal filtering performance, which is neither practical nor time-efficient. Hence, this study proposed to use the metaheuristic Whale Optimization Algorithm (WOA) to estimate the parameters of a simple morphology-based (SMRF) point cloud filtering algorithm to improve its performance, automating the filtering process. The performance of the proposed filtering methodology was compared not only against that of the standard SMRF algorithm but also against those of popular filtering algorithms Cloth Simulation Filtering (CSF) and Progressive TIN Densification (PTIN). The results showed that the proposed filtering methodology outperformed the PTIN and standard SMRF algorithms and presented a comparable performance with the CSF algorithm, which is one of the most robust point cloud filtering algorithms proposed to date. It can also be concluded that metaheuristic optimization algorithms can be used to automate the point cloud filtering process, minimizing the filtering errors caused by user intervention.

Today, the combination of Aerial and Terrestrial photos has been more implemented for 3D modelling purposes. This 3D modelling technique getting popular because it is supporting with photogrammetry structure from motion algorithm (SFM). The SFM algorithm makes automation in the processing step. One of the main problems that will occur in the automation of 3D modelling objects with the SFM algorithm is whether objects have texture-less surfaces. The purpose of this research is to evaluate some enhancement processes that were applied before running the SFM algorithm for 3D modelling. Some pre-processing enhancements are a combination Contrast-Limited Adaptive Histogram EqualizationAlgorithmContrast-Limited Adaptive Histogram Equalization (CLAHE) from Fiji-ImageJ and JPEG to RAW Ai artefact algorithm from Topaz Labs. Two sample objects are tested which are a heritage object that has a texture-less wall surface object and a paddy field that has a similar object pattern. Some aerial and terrestrial photos have been enhanced before processing in 3D modelling. The result shows that applying preprocessing enhancement can improve the completeness of the object, especially in texture-less wall surface area. Pre-processing enhancement improves the geometric accuracy and number of vertex and surfaces also. In the future, the combination of the Jpeg to Raw Ai and the CLAHE enhancement should be explored for the best 3D model solution.

Climate change and local weather conditions have caused several issues in the farming sector. The rapidly expanding global population is an issue that must be addressed to secure food and water supplies through the use of information technology in precision agriculture and smart farming. These technical advances in precision agriculture are represented by unmanned aerial vehicles (UAVs). UAVs or DRONEs help in agriculture by counting the number of plants, visual inspection of the crop field, water management, erosion analysis, plant counting, soil moisture analysis, crop health assessment, irrigation scheduling, analyzing plant physiology, and yield forecasting. Drones can be used to facilitate development by reporting and collecting data in rural development in terms of agriculture land boundaries, water resources and their surface area, village boundaries, monitoring forest area, observation of hilly and tall plant regions, and soil condition in terms of water content, moisture, electrical conductivity, pH, and temperature. Repetitive collection of image and video data helps to analyze changes in rural development. Rural development aims to improve rural communities’ physical infrastructure and basic services. Delay in detecting problems associated with rural development may further deteriorate soil and water resources making them more vulnerable. This paper focuses on various opportunities and challenges in sustainable rural development and the application of UAVs in almost every aspect of human life, allowing people to make significant advances in human life support.

In regions prone to natural disasters, the buildings must follow specific construction standards to avoid demolition. One of the factors that predict the risk of damage is the roof material. This paper investigates the performance of various deep convolutional neural network architectures to classify buildings based on roof material from aerial drone imagery. We also propose a method that is an ensemble of ResNetResNet, ResNeXtResNeXt, and EfficientNetEfficientNet variants of convolutional neural networks, which performed the best in our experiments. We obtained a log loss value as low as 0.4373 using the proposed method. Therefore, the proposed method can be used to perform an accurate classification of roof material using aerial drone imagery.

Remote sensing plays an important role in characterizing the land surface by extensively concerning its spatial resolution. Most of the time, spectral and temporal resolution becomes a limitation, which now can be overcome via unmanned aerial vehicle (UAV) as a remote sensing platform. The study utilizes the google earth engine cloud-based platform to prepare the classified maps from Sentinel 2 and UAV datasets using the Random Forest algorithm. The canopy cover was estimated using UAV data and divided into 4 classes: very dense forest, moderately dense forest, open forest and scrub forest. The majority (39%) areas were under scrub forest. Furthermore, the land use land cover was prepared using UAV data and showed superior results with 95.5% overall accuracy compared with 86.5% of Sentinel 2. Lastly, the tree count of the area was estimated using high-resolution data. The predicted number of trees was 3052, with an accuracy of 82%. The tree count algorithm works better in plantation and even canopy-size trees. Thus, this methodology ultimately helps to achieve the sustainable use of resources concerning their availability, demand and exploitation in the study area. The estimated results can help policymakers, government officials, and local people halt desertification and better sustainable forest management.

Manual power is not sufficient to solve agricultural tasks. Heavy tasks are creating problems of soil contamination and seed contamination. It affected the plant after the locust and plant diseases spread. Drone mapping technology and the classification of DSM ortho mosaic image techniques provided the solutions to the problems. Vegetation indices helped with the identification of plant growth with the help of a drone. Drone mapping and surveys capture hyperspectral images. The images can be calculated using pix4Dmapper. The process is based on initial processing in stage 1, point clouding, meshes generation in stage 2, generation of the index, and DSM and ortho mosaic images in stage 3. We converted 1200 multispectral images and calculated vegetation index values. We measured plant height, plant temperature, the distance between plants, growth vegetation, the soil index of the agricultural land, the water index of the agricultural land, the disease index of the agricultural land, and the vegetation index of the agricultural land. This research proposed identifying the vegetation index on a single agricultural land using an NDVI multispectral image and a hyperspectral image (Geli et al. [1]). We utilized some vegetation indices using drone mapping. The research work started with multispectral image analysis. We collected over 1200 multispectral images in Tif format. It includes NIR band images, Red_edge band images, Green band images, Blue band images, and Red band images. All images are analyzed and tested for calculating vegetation indices of different agricultural land. We have extracted and classified remote sensing images of the agricultural land in a different direction [2]. In the future, we can find the vegetation value of agricultural land and plants using multispectral thermal images for deciding on water irrigation for agricultural places. Our outcome results are displayed as the following: plant growth areas, diseased plant areas, locust damaged plant areas, water detection areas, and soil quality index using the Vegetation index.

The determination of the Dendrometric parameters of forest stands has a silvicultural and ecological interest for the forester, in particular for the evaluation of the dynamics of growth and productivity, and the evaluation of indicators of good ecological status. Currently, UAV-LiDAR (Unmanned Aerial Vehicle-Light Detection and Ranging) has become the new trend for measurement professionals, offering very high-resolution data collection at considerably lower survey costs. In addition, this technology has started to prove its utility in forest inventory applications namely to extract dendrometric parameters, where direct and conventional measurements are sometimes difficult. As for the TLS (Terrestrial Laser Scanning) technology, it has made it possible to obtain several abundant and refined structural information under the forest canopy. In the context of extraction of forest inventory parameters, the precision of extracting tree height for example using TLS alone, is insufficient. Hence the contribution of the combination of ALS (Aerial Laser Scanning) with TLS data to fill any information gaps that may exist. The main goal of this study is to present an approach to the automatic extraction of dendrometric parameters from UAV-LiDAR and TLS data. The proposed methodology is based on performing a TLS survey at a plot level and an ALS scan of the entire area. Our methodology is essentially made up of two steps: automatic crown delineation and automatic extraction of dendrometric parametersdendrometric parameters (position, Diameter at breast height, height, stem curve, concave and convex hull). For the first step, we compared the segmentation of the point cloud by the Watershed algorithm and by the SEGMA pipeline. Whereas the extraction of the dendrometric parameters was carried out using a set of algorithms namely RHT (Random Hough TransformAlgorithmRandom Hough Transform) and LSR (Least Square Regression). The study focused on UAV-ALS and TLS datasets from different regions and with different densities (the Mediterranean, tropical, and coniferous forest). The validation was done using measurements carried out manually on the datasets. The results show that delineation by SEGMA gave a percentage of crown detection varying from 98 to 113% (over-segmentation) with diameters having a coefficient of determination varying from 56 to 90% depending on the area while the Watershed algorithm presented an over-segmentation of the actual crowns. Whereas the results for the DBH determination, the RHT and LSR algorithms both displayed almost 1–4 cm deviations from the reference while the height was extracted with 1–8 mm deviations.

The orchard management has improved by adequately utilizing the Remote Sensing (RS) and Geo-Information System (GIS). With the expansion of orchards in recent years, the sector has been facing a lack of skilled workers and specialists for optimal irrigation utilities, nutrient intake, canopy pruning, pest prevention, disease detection, and orchard quality management name a few. Sustainable orchard management will get a potential boost if the orchard database is well documented with standardized remote observation. Also, the observations recorded have optimal spectral, spatial, and temporal parameters to estimate overall changes in the orchard health. The Unmanned Aerial System (UAS)-based RS and GIS provide an approach that allows users to collect data efficiently and orderly. The following case study focuses on mapping a mango orchard by utilizing the derived product of the UAV remote sensing, an RGB ortho-imageries for analysis and generation of the orchard geo-management system.The geo analysis of the orchard is broadly subdivided into two major categories: spatial and spectral properties of the canopies and surroundings. First, the canopies’ precise position was determined. Next, the tree height is estimated using the shadow’s length, location, and time when the image was captured. From the classified image, individual canopies are labeled, and their top crown size is compared between manually drawn, semi-automatically generated, and field calculated values. It was observed that even though the area difference between these methods was only 1.08 m2 on average, the difference in perimeter was 8.2 m on average. The automated process can precisely map borders to the pixel level. Simultaneously, the manual method is limited to human perception of boundaries and will vary from user to user. The canopies’ spectral response provides insight and permits the interpretation of their physical properties like health, fruit maturity, and diseases. Furthermore, manual and semi-automatic generated canopies were compared concerning object-based averaged spectra. It was observed that the canopies’ histogram was bimodal in the green band in both cases. This can be attributed to the two-year cycle of Mangifera indica. Hence some trees were boring abundant fruits while other canopies bore lesser fruits. Further analysis by estimating canopies center, their accurate position is mapped which is very useful for logistics and management like planning for minimal distance to cover every tree for plucking fruits, visiting, or pruning, simulation of the spread of canopies, simulation of infectious disease, inter canopy gaps (between canopies or ground where sunlight illumination is available) to name a few. Finally, the orchard features such as ‘Canopy Positional Proximity Value’ (CPPV), a positional parameter of an individual tree concerning other trees in the orchard, are defined, which is further used to determine the ‘Orchard Compactness Factor’ (OCF) as an indicator of how densely the trees are packed in an orchard. OFC and CPPV provides information on orchard density as well as shape-size factor for the orchard geo-management. Hence, UAS-RS and GIS are potential tools that can mitigate many problems associated with orchard geo-management, which may further enhance the overall orchard productivity and sustainability.

With the combinational use of geospatial and UAV technology, people have shown that much clearer and more precise surface features of the area under consideration can be extracted. Usually, for this, a good quality camera with limited memory is used. Although there are advancements in battery technology, the amount of time required in extracting data, analyze it and then rescheduling another flight is still a challenge. With advancements in Wi-Fi chips, which are light, reliable, and also cost reasonably less, one can set up a system for the swarm of UAVs to collect geospatial imagery data and simultaneously send that data over to the ground for real-time analysis. This will not only save the time for gathering data but will also provide newer opportunities for research. Also as the overall integrated systems are costly, this technology can be used for smaller missions and tinkered UAV projects. This paper discusses vastly experiments that were done for high-speed data transfer rates, the problem one faces during the design of such systems, and lessons learned for further research. FANETS—Flying Ad-Hoc NETworkS are being widely studied. These days much of the discussions are limited to radio connectivity, which is dependent on heavy equipment loaded on large UAVs. The scope of this study is limited to more affordable, medium to small UAVs that are widely used in geospatial technology as they are agile and small in size. This paper also gives a brief about probabilistic aspects of regular practice that leads to a successful connection. Where there is a large number of nodes, how can hopping help, is also briefly discussed along with its further scopes of research?

Algal blooms are commonly grown in the aquatic environment due to the excessive nutrients (nitrogen and phosphorous) present in ponds and reservoirs. Harmful algal blooms can produce toxic compounds that can contaminate the water and cause serious and harmful effects on aquatic life. Along with algal blooms the presence of other organic matter also makes huge gallons of water unfit for consumption. Remote Sensing is one of the efficient and well-established technologies that is used for the detection of phytoplankton present in the water. Monitoring the rapid growth of algal blooms continuously requires high-resolution spatial and temporal satellite data sets that are costly and hard to get. The images acquired through Unmanned Air Vehicle (UAV) produce high spatial resolution information to continuously monitor the algal bloom growth variation cost-effectively. In this proposed work, the algal bloom presented in the Waddeaplly Lake (Warangal, India) was captured by using a multispectral sensor mounted in DJI Phantom 4 Pro V2.0. RGB images were acquired and pre-processed in Pix4D Mapper Pro to develop the orthomosaic image. NGRDI, NGBDI, GNDVI, and ExGI indices are used in this proposed work to extract the algal bloom matter present in the lake and the data is compared with the Sentinel-2A images for validation purposes. UAV plays an important role in continuously monitoring the algae biomass and developing the precautionary warning system accurately.

The configuration of the camera and Field Of View (FOV) in a closed range photogrammetry are the most important parameters to obtain images of the field in quality. UAVs are central Wide Range photogrammetry tools, with cameras for large field items, for obtaining the optimal standard of photos by offering different usable algorithms of flight planning geographic location. This paper suggests an algorithm to find the exact Geo-Coordinates for the location detection of the UAV flight planning algorithm for the application of Closed Range Photogrammetry to the Battle field Susceptibility Zone analysis. Terrain Slope and Aspect have been considered in order to maintain the measured angle of view, field of vision, and land surface image quality. Identifying the pitch and aspect terrain (Azimuth) using the Digital Elevation Model (DEM) ISRO CartoSat Remote sensing info. In order to improve three-dimensional landscape pictures for 3d modelling, Battle Field Analysis, this algorithm should maximize the functionality of the numerous usable UAV airplane planners.

The rapid advancement of the Unmanned Aerial Systems (UAS) technology and applications provide a unique opportunity to assist the United Nations with their 17 Sustainable Development Goals (SDG’s). Further enhancement of the UAS sector has brought forward tangible applications that illustrate how this technology can assist in improving community health, collective education and stimulate economic growth. There are numerous practical examples of how UAS contribute to the SDG’s of climatologists today. Outlined in this paper are some of the most prominent UAS applications and their resulting benefits to society. The development and global roll out of the UAS has created significant opportunity for climate scientists studying toxic gas emissions and the chemistry of Earth’s atmosphere. UAS make emission monitoring more accurate, contribute to gas leak detection at industrial facilities and greatly assist plant efficiency and optimization. UAS help simplify the data collection process by allowing remote pilots to quickly survey industrial plants and allow them to carry payloads of scientific equipment such as spectrometers which can be used for Differential Optical Absorbance Spectroscopy (DOAS) or thermal cameras designed to detect thermal anomalies. The research illustrated here shows real-world practical examples of using UAS technology on natural emissions at the active crater of the Poás Volcano in Costa Rica and the Industrial power plant of Tampa Electric (TECO) in Tampa, Florida.

In this modern era, uncrewed aerial vehicles (UAVs) have effectively changed and transformed the aeronautical industry in an effective way. A specific technique created to increase this disintegration is the UAV swarm system. The UAV swarm has the ability to deliver missions effectively and self-coordinate the operations of multiple UAVs with no remote operator intervening. This paper comprehensively surveys the literature about various UAV swarm functioning. It proposes a swarm architecture considering the high parameters that allow wireless radio communication infrastructure for a high degree of swarm system reliability and autonomy. This review paper chronicle did preliminary test development to carry out this proposed swarm wireless radio communication system architecture. UAV’s gradual integrated development of UAV swarms with UAV communications Autonomous self-coordination and organizing capability are central to advancing the utility of UAV swarm systems. Several limiting factors hamper the usability of UAVs and reduce performance, including communication, various networking challenges, size-load, and proper power considerations. In addition, the wireless radio system takes advantage of a highly reliable and robust infrastructure for wireless radio communications with secure one machine to another machine.

The network of autonomous Unmanned Aerial Vehicles (UAVs) is a powerful system that can assess the severity of damages during disaster events and support search and rescue missions. UAVs can carry payloads such as cameras, sensors, and a built-in navigation system and can be readily deployed in the surveillance region with limited or no infrastructure support. This work assumes that UAVs can be randomly deployed in the affected area for surveillance. The network is then arranged in the form of clusters of UAV nodes to create a hierarchy and aid in the collection and routing of sensed data. Metrics of residual energy and connectivity have been used to select a Cluster-Head (CH) node iteratively. This proposed clustering algorithm has been detailed in this paper. For the implementation of this protocol, an integrated platform of ROS and NS3 has been utilized to provide a more realistic deployment scenario. The proposed clustering protocol has been compared with prominent clustering protocols of Wireless Sensor Networks (WSNs) such as Hybrid Energy-Efficient Distributed (HEED) and Low-Energy Adaptive Clustering Hierarchy (LEACH) for analysis of parameters such as the lifetime of the network and clustering overhead. The mobility model achieved from the robot simulator has been compared against probabilistic mobility models available in the network simulator. The proposed deterministic clustering protocol outperforms in terms of network lifetime against prominent clustering protocols. Upon stimulation, it has also been observed that the realistic mobility model obtained from the robot simulator is more suited for real-world applications.

Quadcopters take precedence over fixed-wing aircraft within the UAV family, owing to their distinctive characteristics such as vertical take-off and landing, reduced size and weight, high maneuverability, and more. With recent technological advancements, UAVs have become viable for a wide range of applications ranging from military to civilian, including traffic monitoring, aerial photography, surveillance, payload carrying search and rescue, and much more, especially for tedious, filthy, and dangerous jobs that endanger people, such as building fires and observatories in the woods, military purposes. To complete some of these missions, a swarmswarm of Quadcopters must work together. Determining how a Quadcopter can autonomously achieve its goal position despite obstacles in its path is a complex issue. This paper uses the A* (A-star) algorithm to model path planning, trajectory generation, and autonomous control of a quadcopter. Path planning was thoroughly examined using various scenarios with various obstacle positions, and the dimensions of obstacles are much greater than the dimensions of the Quadcopter over the map. It was also observed whether the swarmswarm (3 Quadcopters in “vee” formation) maintains the necessary formation throughout the created path.

Aerial platforms have recently gained significant popularity for the rapid development of relief networks in emergencies. These platforms are capable to deliver essential wireless communication for various applications such as public safety, natural disasters, or adding coverage to existing terrestrial networks. A reliable prediction of coverage resulting from an aerial base station is important to provide essential air-to-ground wireless services for disaster-affected areas. Line-of-sightLine-of-sight (LoS) is an essential component of air-to-ground wireless channels, particularly useful for radio planning and coverage prediction. The performance of an air-to-ground link can be evaluated on three key parameters: elevation angle, communication range, and altitude between the aerial base station and ground receiver. In this paper, we proposed an elevation-dependent line-of-sight model to estimate the area coverage of an aerial base station. The proposed model is derived from statistical parameters of building distribution, defined by the International Telecommunication Union for four urban environments: urban, suburban, dense urban, and high-rise urban. Coverage of aerial base station is estimated from building blockage probability which is formulated as a weighted function of the developed LoS model. Estimated coverage is simulated for elevation angle and communication range between UAV and ground receiver for low altitudes up to 500 m. We restricted UAV altitude up to 500 m due to the limitation on flying altitude by regulating authorities. Our results contribute to identifying the optimum elevation angle and communication range between UAV and ground receiver for line-of-sight communication. Based on the results, we deduced that the optimum elevation angle to attain 100% coverage is between 60 and 80° for all urban environments. We observed a significant reduction in the communication range with declination in UAV altitude, to attain the same amount of coverage for urban, dense urban, and high-rise urban environments. For suburban, altitude is not playing a significant role in the range of communication to achieve area coverage.

Recent emerging technologies are being utilized for the quench of connectivity in real time scenarios. A push is coming to make the information available to humans from the real-time environmental data collected through small sensing devices. Wireless ad-hoc network is a base architecture for Internet of Things (IoT), Unmanned Arial Vehicle (UAV) and drones etc. In this series, IoDTIoDT (Internet of Drone Things) came as the future of drones backend via the Internet of Things, smart vision, cloud computing, enhanced communication, big data, and advanced security approaches. Rapid growth in sensing devices connected to the Internet with intelligence and capabilities also opens the door for attackers because more devices are connected means more chances of security vulnerabilities. Since data authentication is handled only by the central station, which may lead to the chances of device spoofing and false authentication brings less reliability after all. Blockchain (BC) technology is introduced to address such security concerns by eliminating the role of central authority. Blockchain Technology gives decentralized and non-tamperable solutions for the most demanding security service i.e. Authentication. This paper starts with unique characteristics and security challenges in such IOT environment and further covers the authentication process by blockchain with its potential benefits. The multilayer ecosystem is illustrated to fulfill the requirement of the UAV IoT environment, where multiple devices are equally working in a cooperative manner to perform an authorized action. The paper presented a complete study about the integration of blockchain in IoT enabled UAV.

The Drones are used for multifarious activities right from surveillance, express shipping, precision crop monitoring, geographic mapping of inescapable terrain and locations etc. To perform effectively in ibid actions, the drones must keep flying for adequate time in the air. Flying for a longer duration is a necessity for many applications, which is primarily dependent upon the batteries they are using. The batteries which are normally used now-a-days in drones are Lithium Polymer (LiPo) batteries. These batteries are rechargeable and are available in various forms as per the size and use of drones. The longer sustenance of LiPo batteries depends on Voltage level, Capacity, Discharge rateDischarge rate, Activation time, and Charging time. Continuous power to drones is generally maintained by charging aforesaid LiPo batteries whenever Drone returns to earth/site. However, there are other options also to recharge the LiPo batteries, which avoid drones returning to their base. It includes using Polls, Recharging Stations, Solar Voltaic Cells based Drones. Moreover, there is an advanced methodology through which Drones can be recharged using other Drones while in flight. Apropos, this Paper will elaborate upon how to choose an appropriate LiPo Battery Pack based upon various parameters for Drones, Basic principle of working of LiPo batteries (i.e., Intercalation and De-Intercalation), Various forms/Configurations of LiPo batteries and Cell sizes, Maintenance of LiPo batteries and alternates to LiPo batteries for drones which include Charging stations, Charging Drones while on Flight and Use of Solar voltaic cells.

In recent years, as the usage of drones is increased, the batteries of the drones play a vital role in their function. At the same time, though the capacity of the drone battery has been increasing it is not sufficient for many multiple functioning of drones. Moreover, the current technology of drones is used for taking pictures. If there is sufficient energy for drones, they can be used for many multiple functions. In this paper, a new process has been introduced in which the drone batteries are charged through the waves transmitted by the device affixed to a tower, where the tower is used as a source of medium. The receiver chip is affixed to the battery of drones which receives the waves from farther distances and converts them to DC. In this process, series resonance is used in the transmitter device to get the waves and is amplified to get the desired output. The output is transmitted through directional antennas, for overall 36°. These transmitted waves will be received by the receiving chip by filtering the noise and converted to DC (Power supply for batteries of electronic gadgets). The result of this method is that it revolutionizes drone technology because if there are sufficient energy drones can be used for multiple functions.

Plant height is beneficial in defence-related applications during the movement of troops as terrain information is required in advance. This terrain information is of utmost importance to obtain knowledge about possible paths especially in unexplored areas. This information facilitates safe movement of troops. While exploring an unexplored area, vegetation cover area need to be checked carefully, because of the tall and dense bushes. During vegetation monitoring various parameters like plant growth, soil moisture, water availability, the need for fertilizers, etc. are observed. All these parameters are necessarily checked to monitor the growth of the plant. The plant growth parameter is reflected from the plant height. In the modern era, smart farming and precision agriculture have been applied, in which monitoring of plant growth-related parameters are optimized and the necessity of any parameter is fulfilled as per the demand and position in the field. In the extension of smart agriculture, the need for the usage of drones arises while monitoring the fields. Drones provide good spatial resolution and can be flown according to the need and application. In this work, the objective is to calculate the plant height using a drone in order to ensure safe troops movement and on other side to monitor the healthy growth of the plant. The application can be helpful for defense as well as civilian purpose. For achieving this, a machine learning-based model has been proposed in which multiple ground control points (GCPs) of different heights have been used to train the model, which results in minimizing the output error. The challenge is to get drone data and manually recorded GCPs height correctly so that the training can be accomplished successfully for better results.