electrostat · electrostat · May 20, 2025 · May 20, 2025 · May 20, 2025 · May 20, 2025
diff --git a/HyperSpectral/aceScoreCalculation.py b/HyperSpectral/aceScoreCalculation.py
@@ -0,0 +1,19 @@
+import numpy as np
+
+#The Adaptive Cosine Estimator (ACE) algorithm is a popular method used in hyperspectral image processing for target detection.
+#It determines the presence of a specific material within a hyperspectral image by calculating the normalized correlation between the spectrum of each pixel and a known target spectrum.
+
+def aceScoreCalculation(spectrumValA1, spectrumValA2, spectrumValB1, spectrumValB2):
+    # Define the pixel spectrum and target spectrum
+    pixel_spectrum = np.array([spectrumValA1, spectrumValA2])
+    target_spectrum = np.array([spectrumValB1, spectrumValB2])
+
+    # Calculate dot products
+    numerator = np.dot(pixel_spectrum, target_spectrum)**2
+    denominator = np.dot(pixel_spectrum, pixel_spectrum) * np.dot(target_spectrum, target_spectrum)
+
+    # ACE score calculation
+    ace_score = numerator / denominator
+    print(f"ACE Score: {ace_score}")
+
+
diff --git a/HyperSpectral/euclideanDistance.py b/HyperSpectral/euclideanDistance.py
@@ -0,0 +1,10 @@
+import numpy as np
+
+#calculating distance between two points
+#points are np.arrays containing x and y coordinates only
+#must exist within the same "space"
+
+def calcEuclideanDist(x1, x2):
+    ed = np.sqrt(np.matmul(np.matrix.transpose(x1-x2),(x1-x2)))
+    print('The Euclidean distance is: ')
+    print(ed)
diff --git a/HyperSpectral/matrixMath.py b/HyperSpectral/matrixMath.py
@@ -0,0 +1,38 @@
+import numpy as np
+
+#calculating Eigen values, Eigen vectors, and the Coefficient Matrix for a matrix
+def matrixMath(matrix):
+    eigenValues, eigenVectors = np.linalg.eig(matrix)
+
+    print('The eigenvalues of the covariance matrix are: ')
+    print(eigenValues)
+
+    print('The eigenvectors for covariance matrix are: ')
+    print(eigenVectors)
+
+    coefficentMatrix = eigenVectors.transpose()
+    print('The coefficient matrix is: ')
+    print(coefficentMatrix)
+
+    return eigenValues, eigenVectors, coefficentMatrix
+
+#calculate mean of matrix
+def matrixMean(matrix):
+    mean = np.mean(matrix)
+    return mean
+
+#calculate principal components of matrix values and arrays of x,y coordinates
+def principalComponents(matrix, array1, array2):
+    mean = matrixMean(matrix)
+    eigenVal, eigenVectors, coefficientmatrix = matrixMath(matrix)
+
+    pcArray1 = np.matmul(coefficientmatrix, (array1-mean))
+    pcArray2 = np.matmul(coefficientmatrix, (array2-mean))    
+
+    print('The Principal Component transformed value for array1 is: ');
+    print(pcArray1)  
+
+    print('The Principal Component transformed value for array2 is: ');
+    print(pcArray2)
+
+    return pcArray1, pcArray2
diff --git a/HyperSpectral/spectralAngle.py b/HyperSpectral/spectralAngle.py
@@ -0,0 +1,14 @@
+import numpy as np
+
+ # Calculate the angle between two vectors given positions in nonEuclidean space
+def calculateSpectralAngle(spectralArray1, spectralArray2):
+    angle = np.arccos( (np.dot(spectralArray1, spectralArray2)) / (np.linalg.norm(spectralArray1)*(np.linalg.norm(spectralArray2))))
+
+    angleDeg = angle * (180.0/np.pi)
+    print('The Spectral Angle in radians is: ');
+    print(angle)
+
+    print('The Spectral Angle in degrees is: ');
+    print(angleDeg)
+
+    return angle, angleDeg