The main theme of this study is to analyze the Soviet economic theoretical debate in the period 1985 – 1991. This period of reconstruction gave possibilities of a more free debate. In the period up to 1989/90 the directive from the Central Committee of the Communist Party was to defend the socialist economic system and its supremacy over market economics. However, certain market economic ideas were deemed as functioning methods also in a planned economic system. One of the conclusions in this thesis is that the Soviet economists failed to solve some central theoretical problems in the Soviet economy and as consequence their thinking failed to have a stabilizing effect on the socialist economic theory. The Achilles heel was how to apply the labour theory of value on a planned economy. In 1990 and 1991 the discussion was very free and now a transition to market economy was accepted by the economists. The main issue between the Soviet economists became now whether a gradual transition to market economy was to be preferred to shock therapy. The majority of the economists recommended a gradual transition. Scholars have emphasized that old stationary structures are important in Russian and Soviet history. A conclusion in this thesis is that such structures seemed to have played a role in Soviet and Russian theoretical thinking in the period 1985 – 1991.

In this thesis, a mixed type video segmentation algorithm is implemented to find the scene cuts in MPEG compressed video data. The main aim is to have a computationally efficient algorithm for real time applications. Due to this reason partial decoding of the bitstream is used in segmentation. As a result of partial decoding, features such as bitrate, motion vector type, and DC images are implemented to find both continuous and discontinuous scene cuts on a MPEG-2 coded general TV broadcast data. The results are also compared with techniques found in literature.

Les technologies multimédias ont récemment connues une grande évolution surtout avec la croissance rapide d'internet ainsi que la création quotidienne de grands volumes de données vidéos. Tout ceci nécessite de nouvelles méthodes performantes permettant d'indexer, de naviguer, de rechercher et de consulter les informations stockées dans de grandes bases de données multimédia. La récupération de données basée sur le contenu vidéo, qui est devenue un domaine de recherche très actif durant cette décennie, regroupe les différentes techniques conçues pour le traitement de la vidéo. Dans le cadre de cette thèse de doctorat, nous présentons des applications permettant la segmentation temporelle d'une vidéo ainsi que la récupération d'information pertinente dans une séquence vidéo. Une fois le processus de classification effectué, il devient possible de rechercher l'information utile en ajoutant de nouveaux critères, et aussi de visualiser l'information d'une manière appropriée permettant d'optimiser le temps et la mémoire. Dans une séquence vidéo, le plan est considéré comme l'unité élémentaire de la vidéo. Un plan est défini comme une suite d'image capturée par une même caméra représentant une action dans le temps. Pour composer une vidéo, plusieurs plans sont regroupés en utilisant des séquences de transitions. Ces transitions se catégorisent en transitions brusques et transitions progressives. Détecter les transitions présentes dans une séquence vidéo a fait l'objet de nos premières recherches. Plusieurs techniques, basées sur différents modèles mathématiques, ont été élaborées pour la détection des changements de plans. L'utilisation de la décomposition en valeur singulière (SVD) ains que la norme Frobenius ont permis d'obtenir des résultats précis en un temps de calcul réduit. Le résumé automatique des séquences vidéo est actuellement un sujet d'une très grande actualité. Comme son nom l'indique, il s'agit d'une version courte de la vidéo qui doit contenir l'essentiel de l'information, tout en étant le plus concis possible. Ils existent deux grandes familles de résumé : le résumé statique et le résumé dynamique. Sélectionner une image représentative de chaque plan permet de créer un scénarimage. Ceci est considéré comme étant un résumé statique et local. Dans notre travail, une méthode de résumé globale est proposée. / With the recent advancement in multimedia technologies, in conjunction with the rapid increase of the volume of digital video data and the growth of internet ; it has becom mandatory to have the hability browse and search through information stored in large multimedia databases. For this purpose, content based video retrieval (CBVR) has become an active area of research durinf the last decade. The objective of this thesis is to present applications for temporal video segmentation and video retrieval based on different mathematical models. A shot is considered as the elementary unit of a video, and is defined as a continuous sequence of frames taken from a single camera, representing an action during time. The different types of transitions that may occur in a video sequence are categorized into : abrupt and gradual transition. In this work, through statistical analysis, we segment a video into its constituent units. This is achieved by identifying transitions between adjacent shots. The first proposed algorithm aims to detect abrupt shot transitions only by measuring the similarity between consecutive frames. Given the size of the vector containing distances, it can be modeled by a log normal distribution since all the values are positive. Gradual shot transition identification is a more difficult task when compared to cut detection. Generally, a gradual transition may share similar characteristics as a dynamic segment with camera or object motion. In this work, singular value decomposition (SVD) is performed to project features from the spatial domain to the singular space. Resulting features are reduced and more refined, which makes the remaining tasks easier. The proposed system, designed for detecting both abrupt and gradual transitions, has lead to reliable performances achieving high detection rates. In addition, the acceptable computational time allows to process in real time. Once a video is partitioned into its elementary units, high-level applications can be processed, such as the key-frame extraction. Selecting representative frames from each shot to form a storyboard is considered as a static and local video summarization. In our research, we opted for a global method based on local extraction. Using refined centrist features from the singular space, we select representative frames using modified k-means clustering based on important scenes. This leads to catch pertinent frames without redoudancy in the final storyboard.