Automated Urban Travel Interpretation: A Bottom-up Approach for Trajectory Segmentation

View full-text article in PMC

. 2016 Nov 23;16(11):1962. doi: 10.3390/s16111962

Algorithm 4 Rules for segment merging

1:
INPUT All the atomic segments in a given trajectory (seglist), temporal threshold: Ψ
2:
OUPUT a set of merged segments (mergedSeglist)
3:
PROCEDURE Segment_Merging()
4:
$s e g l i s t . s i z e () = k$
5:
boolean $f l a g = f a l s e$
6:
for i=0 to k−2 do
7:
$c u r r e n t_s e g = s e g l i s t . g e t (i)$
8:
$n e x t_s e g = s e g l i s t . g e t (i + 1)$
9:
RULE 1:
10:
if $c u r r e n t_s e g . m o d e_t y p e = = n e x t_s e g . m o d e_t y p e & & n e x t_s e g . d u r a t i o n \leq Ψ$ then
11:
$m e r g e_s e g = m e r g i n g (c u r r e n t_s e g, n e x t_s e g)$
12:
$m e r g e d S e g l i s t . a d d (m e r g e_s e g)$
13:
$f l a g = t r u e$
14:
end if
15:
RULE 2:
16:
if $c u r r e n t_s e g . m o d e_t y p e = = n e x t_s e g . m o d e_t y p e & & n e x t_s e g . d u r a t i o n > Ψ$ then
17:
$m e r g e_s e g = m e r g i n g (c u r r e n t_s e g, n e x t_s e g)$
18:
$m e r g e d S e g l i s t . a d d (m e r g e_s e g)$
19:
$f l a g = t r u e$
20:
end if
21:
RULE 3:
22:
if $c u r r e n t_s e g . m o d e_t y p e! = n e x t_s e g . m o d e_t y p e & & n e x t_s e g . d u r a t i o n \leq Ψ$ then
23:
$m e r g e_s e g = m e r g i n g (c u r r e n t_s e g, n e x t_s e g)$
24:
$m e r g e d S e g l i s t . a d d (m e r g e_s e g)$
25:
$f l a g = t r u e$
26:
end if
27:
RULE 4:
28:
if $f l a g = = f a l s e$ then
29:
$m e r g e d S e g l i s t . a d d (c u r r e n t_s e g)$
30:
$m e r g e d S e g l i s t . a d d (n e x t_s e g)$
31:
end if
32:
end for
33:
END PROCEDURE