Method to convert the raw signals into images. First, an M 2 signal sub-sample is taken, where M represents the total height and width of the square image. This sub-sample is then mapped into a matrix and each point is normalized in a range from 0 to 255 to represent the intensity of each pixel value.

Contexts in source publication

Context 1

... that converts raw vibration signals in time-domain to images, in order to take advantage of the powerful classification tools available for image processing using CNN [12,23,18,21,22,27,42,43]. Moreover, converting the raw signals into images provides a good way to explore two-dimensional features [42]. For a raw signal R with N sample points, Fig. 2 shows the conversion method to images, where each time-domain signal point is one pixel of a square grayscale image with size M × M. First a sub-sample L of M 2 size is taken from the raw signal R, hence, the i-th sub-sample is given by L i = {R(i·s+1), R(i·s+2), ..., R(i·s+M 2 )}, where s ∈ Z + is the step between samples, and the ...

Context 2

... signal point is one pixel of a square grayscale image with size M × M. First a sub-sample L of M 2 size is taken from the raw signal R, hence, the i-th sub-sample is given by L i = {R(i·s+1), R(i·s+2), ..., R(i·s+M 2 )}, where s ∈ Z + is the step between samples, and the index i = {0, 1, ..., N/s}, with denoting the floor function (see Fig. 2). Note that we aim for an important overlap between samples in order to obtain more images, which is advantageous for the CNN classifier, i.e. s << M 2 . Then, each point in the sub-sample L fills a matrix of M × M from left to right and from top to bottom. Each point is normalized from 0 to 255, and represents the grayscale intensity ...

Context 3

... that converts raw vibration signals in time-domain to images, in order to take advantage of the powerful classification tools available for image processing using CNN [12,23,18,21,22,27,42,43]. Moreover, converting the raw signals into images provides a good way to explore two-dimensional features [42]. For a raw signal R with N sample points, Fig. 2 shows the conversion method to images, where each time-domain signal point is one pixel of a square grayscale image with size M × M. First a sub-sample L of M 2 size is taken from the raw signal R, hence, the i-th sub-sample is given by L i = {R(i·s+1), R(i·s+2), ..., R(i·s+M 2 )}, where s ∈ Z + is the step between samples, and the ...

Context 4

... signal point is one pixel of a square grayscale image with size M × M. First a sub-sample L of M 2 size is taken from the raw signal R, hence, the i-th sub-sample is given by L i = {R(i·s+1), R(i·s+2), ..., R(i·s+M 2 )}, where s ∈ Z + is the step between samples, and the index i = {0, 1, ..., N/s}, with denoting the floor function (see Fig. 2). Note that we aim for an important overlap between samples in order to obtain more images, which is advantageous for the CNN classifier, i.e. s << M 2 . Then, each point in the sub-sample L fills a matrix of M × M from left to right and from top to bottom. Each point is normalized from 0 to 255, and represents the grayscale intensity ...