I am trying to segment words in a handwritten text line. I am doing this based on a research paper whose word segmentation part is given in the image. I do not understand the quantities for which the histogram is to be made.Histogram for word segmentation-(image link)
This paper is written is some "easy-go" fashion.
But lets try to understand.
V() is not a histogram but a filtering mechanism: you work down the line on each column and find the average. I assume the image I(i, j) is binary/black-white (or maybe gray-scale).
V(i)=Sum_j I(i, j)/height
is the average of image intensities on column i.
So you just calculate these V(i) for i=0, .., width-1.
Then you go from there.