Earth Engine supporta le trasformazioni degli array, come la trasposizione, l'inverso e lo pseudo-inverso. Ad esempio, prendiamo in considerazione una regressione con minimi quadrati ordinari (OLS) di una serie temporale di immagini. Nell'esempio seguente, un'immagine con bande per gli indicatori e una risposta viene convertita in un'immagine array, quindi "risoluta" per ottenere le stime dei coefficienti minimi quadratici in tre modi. Innanzitutto, assembla i dati dell'immagine e convertili in array:
Editor di codice (JavaScript)
// Scales and masks Landsat 8 surface reflectance images. function prepSrL8(image) { // Develop masks for unwanted pixels (fill, cloud, cloud shadow). var qaMask = image.select('QA_PIXEL').bitwiseAnd(parseInt('11111', 2)).eq(0); var saturationMask = image.select('QA_RADSAT').eq(0); // Apply the scaling factors to the appropriate bands. var opticalBands = image.select('SR_B.').multiply(0.0000275).add(-0.2); var thermalBands = image.select('ST_B.*').multiply(0.00341802).add(149.0); // Replace the original bands with the scaled ones and apply the masks. return image.addBands(opticalBands, null, true) .addBands(thermalBands, null, true) .updateMask(qaMask) .updateMask(saturationMask); } // Load a Landsat 8 surface reflectance image collection. var collection = ee.ImageCollection('LANDSAT/LC08/C02/T1_L2') // Filter to get only two years of data. .filterDate('2019-04-01', '2021-04-01') // Filter to get only imagery at a point of interest. .filterBounds(ee.Geometry.Point(-122.08709, 36.9732)) // Prepare images by mapping the prepSrL8 function over the collection. .map(prepSrL8) // Select NIR and red bands only. .select(['SR_B5', 'SR_B4']) // Sort the collection in chronological order. .sort('system:time_start', true); // This function computes the predictors and the response from the input. var makeVariables = function(image) { // Compute time of the image in fractional years relative to the Epoch. var year = ee.Image(image.date().difference(ee.Date('1970-01-01'), 'year')); // Compute the season in radians, one cycle per year. var season = year.multiply(2 * Math.PI); // Return an image of the predictors followed by the response. return image.select() .addBands(ee.Image(1)) // 0. constant .addBands(year.rename('t')) // 1. linear trend .addBands(season.sin().rename('sin')) // 2. seasonal .addBands(season.cos().rename('cos')) // 3. seasonal .addBands(image.normalizedDifference().rename('NDVI')) // 4. response .toFloat(); }; // Define the axes of variation in the collection array. var imageAxis = 0; var bandAxis = 1; // Convert the collection to an array. var array = collection.map(makeVariables).toArray(); // Check the length of the image axis (number of images). var arrayLength = array.arrayLength(imageAxis); // Update the mask to ensure that the number of images is greater than or // equal to the number of predictors (the linear model is solvable). array = array.updateMask(arrayLength.gt(4)); // Get slices of the array according to positions along the band axis. var predictors = array.arraySlice(bandAxis, 0, 4); var response = array.arraySlice(bandAxis, 4);
import ee import geemap.core as geemap
Colab (Python)
import math # Scales and masks Landsat 8 surface reflectance images. def prep_sr_l8(image): # Develop masks for unwanted pixels (fill, cloud, cloud shadow). qa_mask = image.select('QA_PIXEL').bitwiseAnd(int('11111', 2)).eq(0) saturation_mask = image.select('QA_RADSAT').eq(0) # Apply the scaling factors to the appropriate bands. optical_bands = image.select('SR_B.').multiply(0.0000275).add(-0.2) thermal_bands = image.select('ST_B.*').multiply(0.00341802).add(149.0) # Replace the original bands with the scaled ones and apply the masks. return ( image.addBands(optical_bands, None, True) .addBands(thermal_bands, None, True) .updateMask(qa_mask) .updateMask(saturation_mask) ) # Load a Landsat 8 surface reflectance image collection. collection = ( ee.ImageCollection('LANDSAT/LC08/C02/T1_L2') # Filter to get only two years of data. .filterDate('2019-04-01', '2021-04-01') # Filter to get only imagery at a point of interest. .filterBounds(ee.Geometry.Point(-122.08709, 36.9732)) # Prepare images by mapping the prep_sr_l8 function over the collection. .map(prep_sr_l8) # Select NIR and red bands only. .select(['SR_B5', 'SR_B4']) # Sort the collection in chronological order. .sort('system:time_start', True) ) # This function computes the predictors and the response from the input. def make_variables(image): # Compute time of the image in fractional years relative to the Epoch. year = ee.Image(image.date().difference(ee.Date('1970-01-01'), 'year')) # Compute the season in radians, one cycle per year. season = year.multiply(2 * math.pi) # Return an image of the predictors followed by the response. return ( image.select() .addBands(ee.Image(1)) # 0. constant .addBands(year.rename('t')) # 1. linear trend .addBands(season.sin().rename('sin')) # 2. seasonal .addBands(season.cos().rename('cos')) # 3. seasonal .addBands(image.normalizedDifference().rename('NDVI')) # 4. response .toFloat() ) # Define the axes of variation in the collection array. image_axis = 0 band_axis = 1 # Convert the collection to an array. array = collection.map(make_variables).toArray() # Check the length of the image axis (number of images). array_length = array.arrayLength(image_axis) # Update the mask to ensure that the number of images is greater than or # equal to the number of predictors (the linear model is solvable). array = array.updateMask(array_length.gt(4)) # Get slices of the array according to positions along the band axis. predictors = array.arraySlice(band_axis, 0, 4) response = array.arraySlice(band_axis, 4)
Tieni presente che arraySlice()
restituisce tutte le immagini nella serie temporale per l'intervallo di indici specificato lungo bandAxis
(l'asse 1). A questo punto,
è possibile utilizzare l'algebra matriciale per risolvere i coefficienti OLS:
Editor di codice (JavaScript)
// Compute coefficients the hard way. var coefficients1 = predictors.arrayTranspose().matrixMultiply(predictors) .matrixInverse().matrixMultiply(predictors.arrayTranspose()) .matrixMultiply(response);
import ee import geemap.core as geemap
Colab (Python)
# Compute coefficients the hard way. coefficients_1 = ( predictors.arrayTranspose() .matrixMultiply(predictors) .matrixInverse() .matrixMultiply(predictors.arrayTranspose()) .matrixMultiply(response) )
Anche se questo metodo funziona, non è efficiente e rende difficile la lettura del codice. Un modo migliore è utilizzare il metodo pseudoInverse()
(matrixPseudoInverse()
per un'immagine array):
Editor di codice (JavaScript)
// Compute coefficients the easy way. var coefficients2 = predictors.matrixPseudoInverse() .matrixMultiply(response);
import ee import geemap.core as geemap
Colab (Python)
# Compute coefficients the easy way. coefficients_2 = predictors.matrixPseudoInverse().matrixMultiply(response)
Dal punto di vista della leggibilità e dell'efficienza di calcolo, il modo migliore per ottenere i coefficienti OLS è solve()
(matrixSolve()
per un'immagine array). La funzione
solve()
determina come risolvere al meglio il sistema dalle caratteristiche
degli input, utilizzando l'operatore pseudo-inverso per i sistemi sovradeterminati, l'operatore inverso per le matrici quadrate e tecniche speciali per le matrici quasi singolari:
Editor di codice (JavaScript)
// Compute coefficients the easiest way. var coefficients3 = predictors.matrixSolve(response);
import ee import geemap.core as geemap
Colab (Python)
# Compute coefficients the easiest way. coefficients_3 = predictors.matrixSolve(response)
Per ottenere un'immagine multibanda, proietta l'immagine dell'array in uno spazio di dimensioni inferiori, quindi appiattiscila:
Editor di codice (JavaScript)
// Turn the results into a multi-band image. var coefficientsImage = coefficients3 // Get rid of the extra dimensions. .arrayProject([0]) .arrayFlatten([ ['constant', 'trend', 'sin', 'cos'] ]);
import ee import geemap.core as geemap
Colab (Python)
# Turn the results into a multi-band image. coefficients_image = ( coefficients_3 # Get rid of the extra dimensions. .arrayProject([0]).arrayFlatten([['constant', 'trend', 'sin', 'cos']]) )
Esamina gli output dei tre metodi e osserva che la matrice risultante dei coefficienti è la stessa indipendentemente dal risolutore. Il fatto che solve()
sia flessibile
ed efficiente lo rende una buona scelta per la creazione di modelli lineari generici.