CDF	Cumulative Distribution Function: this function of image processing presents the resulting image as a linear cumulative distribution function.
CED	Canny Edge Detection: this edge detection operator uses a multi-stage algorithm to detect a wide range of edges in images. It locates the intensity gradients of the image and applies non-maximum suppression remove spurious response in the edges.
CNN	Convolutional Neural Network: this provides the function of classification of images.
ED	Edge Detection: this technique identifies points in a digital image with discontinuities. It sharpens changes in the image brightness.
FN	False Negative: this provides the predicted “no”, indicating that non-defective concrete images are classified as “defective”.
FP	False Positive: this provides the predicted “yes”, indicating that cracked images are classified inaccurately as “non-defective”.
HE	Histogram Equalization: this function is a method in digital image processing that provides contrast adjustment using the histogram of the sampled image.
LOG-ED	Laplacian of Gaussian Edge Detection: initially, this smoothens an image, and it then calculates the Laplacian. The process results in a double-edged image. It finds edges and then locates the zero-crossing between the double edges.
ML	Machine Learning: this simply predicts outcomes of classifying the sampled images. Machine learning algorithms use historical data as input to predict new output values.
PED	Prewitt Edge Detection: this operator is appropriate for detecting the magnitude and orientation of edges. It also has the same parameters as Sobel edge detection; however, it is easier to implement.
RED	Robert’s Edge Detection: this operator is a straightforward and efficient approach to quantifying an image’s spatial gradient. The pixel value at a location in the produced image represents the estimated absolute magnitude value of the inputted image’s spatial gradient at that location.
SED	Sobel Edge Detection: this operator works by calculating the gradient of the intensity of the digital image at each pixel within the image. It locates the direction of the maximum increase from light to dark and the rate of change in that direction.
SNR	Signal-to-Noise Ratio: this function is a general metric for determining image quality. It is described as the relative strength of an aimed signal from a sample compared with the undesired background signal from noise.
TN	True Negative: this provides the predicted “no”, indicating that cracked images are classified correctly as “defective”.
TP	True Positive: this provides the predicted “yes”, indicating that non-defective concrete images are classified correctly as “non-defective”.