No abstract

In this paper we present a new 3-D laparoscopic device based on structured light for minimally invasive surgery. Real-time reconstruction of internal organs' surfaces is very challenging as the numerous geometric and photometric variabilities and disturbances (bloody parts, specularities, smokes,...) often occur during the surgical operation, sometimes with manipulations by several assistants. We then conceived a structured light vision system to illuminate a coded pattern by means of an external video projector device or miniaturized diffractive optical elements and a laser source. Among the structured light techniques, the spatial neighbourhood scheme is the most relevant class of approaches to deal with moving and deformable surfaces, then to capture the depth map with only one shot. Each neighbourhood (a (3 × 3) window) is representing a codeword of length 9, and is unique in the whole pattern, even if there is a lack of information. To do so, a monochromatic subperfect map-based pattern is computed, driven by a desired minimal Hamming distance, H(min), between any couple of codewords. This provides patterns with high correction capabilities (H(min) > 1). For practical considerations, each numerical codeword symbol is associated to a unique visual feature embedding the local orientation of the pattern, which is helpful for the neighbourhood retrieval during the decoding process. Together with the endoscopic device, in vivo real-time reconstructions (in mini-invasive surgical conditions) are presented to assess both the efficiency of the proposed pattern design, the decoding process and the 3-D laparoscope setup realized in the lab.

No abstract

No abstract