A light stripe vision system is used to measure the location of polyhedral features of parts from a single frame of video camera output. Issues such as accuracy in locating the line segments of intersection in the image and combining redundant information from multiple measurements and multiple sources are addressed. In 2.5 seconds, a prototype sensor was capable of locating a two inch cube to an accuracy (one standard deviation) of .002 inches (.055 mm) in translation and .1 degrees (.0015 radians) in rotation. When integrated with a manipulator, the system was capable of performing high precision assembly tasks.

Real-time inspection based on machine vision technologies is being widely used in quality control and cost reduction in a variety of application domains. The high demands on the inspection performance and low cost requirements make the algorithm design a challenging task that requires new and innovative methodologies in image processing and fusion. In this research, an integrated approach that combines novel image processing and fusion techniques is proposed for the efficient design of accurate and real-time machine vision-based inspection algorithms with an application to the food processing problem. Firstly, a general methodology is introduced for effective detection of defects and foreign objects that possess certain spectral and shape features. The factors that affect performance metrics are analyzed, and a recursive segmentation and classification scheme is proposed in order to improve the segmentation accuracy. The developed methodology is applied to real-time fan bone detection in deboned poultry meat with a detection rate of 93% and a false alarm rate of 7% from a lab-scale testing on 280 samples. Secondly, a novel snake-based algorithm is developed for the segmentation of vector-valued images. The snakes are driven by the weighted sum of the optimal forces derived from corresponding energy functionals in each image, where the weights are determined based on a novel metric that measures both local contrasts and noise powers in individual sensor images. This algorithm is effective in improving the segmentation accuracy when imagery from multiple sensors is available to the inspection system. The effectiveness of the developed algorithm is verified using (i) synthesized images (ii) real medical and aerial images and (iii) color and x-ray chicken breast images. The results further confirmed that the algorithm yields higher segmentation accuracy than monosensory methods and is able to accommodate a certain amount of registration error. This feature-level image fusion technique can be combined with pixel- and decision- level techniques to improve the overall inspection system performance.

Cette thèse a pour but de contribuer à améliorer les interfaces Homme-machine. En particulier, nos appareils devraient répliquer notre capacité à traiter continûment des flux d'information. Cependant, le domaine de l’apprentissage statistique dédié à la reconnaissance de séries temporelles pose de multiples défis. Nos travaux utilisent la reconnaissance de gestes comme exemple applicatif, ces données offrent un mélange complexe de poses corporelles et de mouvements, encodées sous des modalités très variées. La première partie de notre travail compare deux modèles temporels de l’état de l’art pour la reconnaissance continue sur des séquences, plus précisément l’hybride réseau de neurones -- modèle de Markov caché (NN-HMM) et les réseaux de neurones récurrents bidirectionnels (BD-RNN) avec des unités commandées par des portes. Pour ce faire, nous avons implémenté un environnement de test partagé qui est plus favorable à une étude comparative équitable. Nous proposons des ajustements sur les fonctions de coût utilisées pour entraîner les réseaux de neurones et sur les expressions du modèle hybride afin de gérer un large déséquilibre des classes de notre base d’apprentissage. Bien que les publications récentes semblent privilégier l’architecture BD-RNN, nous démontrons que l’hybride NN-HMM demeure compétitif. Cependant, ce dernier est plus dépendant de son modèle d'entrées pour modéliser les phénomènes temporels à court terme. Enfin, nous montrons que les facteurs de variations appris sur les entrées par les deux modèles sont inter-compatibles. Dans un second temps, nous présentons une étude de l'apprentissage dit «en un coup» appliqué aux gestes. Ce paradigme d'apprentissage gagne en attention mais demeure peu abordé dans le cas de séries temporelles. Nous proposons une architecture construite autour d’un réseau de neurones bidirectionnel. Son efficacité est démontrée par la reconnaissance de gestes isolés issus d’un dictionnaire de langage des signes. À partir de ce modèle de référence, nous proposons de multiples améliorations inspirées par des travaux dans des domaines connexes, et nous étudions les avantages ou inconvénients de chacun / This thesis aims to improve the intuitiveness of human-computer interfaces. In particular, machines should try to replicate human's ability to process streams of information continuously. However, the sub-domain of Machine Learning dedicated to recognition on time series remains barred by numerous challenges. Our studies use gesture recognition as an exemplar application, gestures intermix static body poses and movements in a complex manner using widely different modalities. The first part of our work compares two state-of-the-art temporal models for continuous sequence recognition, namely Hybrid Neural Network--Hidden Markov Models (NN-HMM) and Bidirectional Recurrent Neural Networks (BDRNN) with gated units. To do so, we reimplemented the two within a shared test-bed which is more amenable to a fair comparative work. We propose adjustments to Neural Network training losses and the Hybrid NN-HMM expressions to accommodate for highly imbalanced data classes. Although recent publications tend to prefer BDRNNs, we demonstrate that Hybrid NN-HMM remain competitive. However, the latter rely significantly on their input layers to model short-term patterns. Finally, we show that input representations learned via both approaches are largely inter-compatible. The second part of our work studies one-shot learning, which has received relatively little attention so far, in particular for sequential inputs such as gestures. We propose a model built around a Bidirectional Recurrent Neural Network. Its effectiveness is demonstrated on the recognition of isolated gestures from a sign language lexicon. We propose several improvements over this baseline by drawing inspiration from related works and evaluate their performances, exhibiting different advantages and disadvantages for each