RESUMEN
The present paper proposes an implementation of a hybrid hardware-software system for the visual servoing of prosthetic arms. We focus on the most critical vision analysis part of the system. The prosthetic system comprises a glass-worn eye tracker and a video camera, and the task is to recognize the object to grasp. The lightweight architecture for gaze-driven object recognition has to be implemented as a wearable device with low power consumption (less than 5.6 W). The algorithmic chain comprises gaze fixations estimation and filtering, generation of candidates, and recognition, with two backbone convolutional neural networks (CNN). The time-consuming parts of the system, such as SIFT (Scale Invariant Feature Transform) detector and the backbone CNN feature extractor, are implemented in FPGA, and a new reduction layer is introduced in the object-recognition CNN to reduce the computational burden. The proposed implementation is compatible with the real-time control of the prosthetic arm.
RESUMEN
This paper presents new methods to study the shape of tubular organs. Determining precise cross-sections is of major importance to perform geometrical measurements, such as diameter, wall-thickness estimation or area measurement. Our first contribution is a robust method to estimate orthogonal planes based on the Voronoi Covariance Measure. Our method is not relying on a curve-skeleton computation beforehand. This means our orthogonal plane estimator can be used either on the skeleton or on the volume. Another important step towards tubular organ characterization is achieved through curve-skeletonization, as skeletons allow to compare two tubular organs, and to perform virtual endoscopy. Our second contribution is dedicated to correcting common defects of the skeleton by new pruning and recentering methods. Finally, we propose a new method for curve-skeleton extraction. Various results are shown on different types of segmented tubular organs, such as neurons, airway-tree and blood vessels.