To detect the boundaries of the segments, we may use [some] text and visual signals as we scan the page top-down or bottom-up.
By "we" I mean "we as human"
Do I need that "some"? Because I feel, using "some" I mean a number of them are used in detecting the segment boundaries. However, if I don't use it, I may imply a generic term for text and visual signals. Right? Which is more usual? I feel English tend to use generic nouns in such cases.