Neural Network

Do you want to see more videos like this? Then subscribe and turn on notifications! Don't forget to subscribe to my YouTube channel and RuTube channel. Rutube : https://rutube.ru/video/914636e7b4190a76fae09e1e91c2911e/ This program facilitates coordinate transformation between two 3D geodetic systems by modeling the differences in X, Y, and Z coordinates using three distinct mathematical approaches: a backpropagation neural network (BPNN), Helmert transformation, and Affine transformation. The transformation is achieved by mapping input coordinates from one system to target coordinates in another, capturing both linear and nonlinear relationships. The BPNN, a flexible nonlinear model, learns complex transformations through a configurable architecture, including a hidden layer with adjustable neuron counts, learning rate, and regularization to prevent overfitting. The Helmert transformation, a rigid-body model, estimates seven parameters: translations along X, Y, Z axes, rotations (expressed as Euler angles: Roll, Pitch, Yaw), and a uniform scale factor, derived via least-squares optimization to align the two systems. The Affine transformation extends this by estimating a 3x4 matrix that accounts for translations, rotations, scaling, and shearing per axis, also computed using least-squares. To enhance numerical stability for large coordinate values typical in geodetic datasets, the program applies mean-centering and min-max normalization during data preprocessing. Model performance is rigorously evaluated using K-Fold cross-validation (Cross-Validation) and bootstrap resampling to estimate the confidence interval of the Root Mean Square Error (RMSE). The program outputs comprehensive metrics, including RMSE, Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Coefficient of Determination (R²), Theil's U Statistic, Nash-Sutcliffe Efficiency (NSE), Kling-Gupta Efficiency (KGE), and geometric measures such as average 3D distance. It also provides per-axis error analysis (mean, standard deviation, skewness, kurtosis), regional accuracy by quadrants, and training statistics (epochs to convergence, final loss, convergence rate). For the Helmert transformation, it reports translation, scale, and rotation parameters, including Euler angles (Roll, Pitch, Yaw) in degrees. This program is ideal for geodetic applications requiring precise coordinate transformations, such as aligning local survey data with global reference systems, correcting systematic distortions in Global Navigation Satellite System (GNSS) surveys, or transforming historical geodetic datasets. It is particularly suited for cadastral mapping, deformation analysis, geospatial data integration, and scenarios involving nonlinear transformations or small datasets with significant coordinate variations. RuTube Channel : https://rutube.ru/channel/1400643/ Youtube Channel : https://www.youtube.com/channel/UCNnd6dzXmsw7MeC1QlP4C4A LinkeDin : https://www.linkedin.com/in/sasalazic/ SrbsBuk : https://srbsbuk.com/members-2/donlaki/ Telegram : https://t.me/geoalatiplus web pages: http://geoalatiplus.in.rs #gnss #gps #agro #agriculture #latitude #longitude #geodetic #geodesy #surveyor #survey #surveying #srbija #projection #ellipsoid #etrs #etrf #itrs #itrf #landsurveyor #geo #геодезия #геодезија #geodezija #WGS #pravougle #koordinate #cors #latitude #longitude #agros #rgz #GNSS #PPP #RTK #Navigacija #Inzenjerstvo #GLONASS #GPS #Galileo #BeiDou #NavIC #NaukaITehnologija