Improved stripe matching for colour encoded structured light

Pattern codification strategies in structured light systems

“…Monks et al [31] designed a pattern based on horizontal coloured stripes in order to reconstruct dynamic scenes. A total number of 6 colours were used to paint the slits, separated by black bands.…”

Structured light camera calibration

“…the plane of light rays which enlight a curve G on a 3D object's surface (Fig. 8) is applied [19][20][21][22][23][24][25][26][27] …”

An overview of the advantages and constraints of coded pattern projection techniques for autonomous navigation

“…This work surveys several techniques which codify the pattern projected on the scene. The advantages and disadvantages have been discussed analysing the following capabilities: a) to measure dynamic scenes, where only one pattern can be projected [15,27,[31][32][33][34][35][36][37][38], and static scenes, where the sequential projection of several patterns is allowed [16-26, 28-30, 391, b) to measure scenes made by highly saturated colour objects, with basically only binary patterns can be projected [15-331; or to measure scenes with a colour content predominantly neutral, where the emission of colour is permitted [34,[36][37][38][39][40][41][42]; specular or metallic surfaces should also be included, but they always lead to a specific system [44,451. The reflection of a pattern region on another surface, already illuminated, produces an evident identification error of the image captured by the camera; c) to measure scenes with a lot of discontinuities, of several different depths, where an absolute codification is only allowed [16-26, 28, 30-421; or to measure scenes predominantly continuous [27], where very easy periodical patterns can be projected [29].…”

“…[38] assert that the matching process proposed by Boyer it is not optimal, because if the matches for the first vertical slit contain errors, these will be propagated through the rest of the data. Monks et al have developed a new technique which eliminates this problem and has the advantage of being non-iterative, encoding the topology of the entire set of stripes as a directed acyclic graph.…”

“…The pattern proposed by Boyer et al [37] and Monks et al [38], is limited to measuring predominantly neutral colour surfaces, as highly saturated hues could produce slit identification errors. Even so, it is specially recommended in dynamic environments as only one projection is needed.…”