Veículos Aéreos não tripulados (VANTs) vêm sendo cada vez mais utilizados em diversos países, tanto na área militar como na civil. O cenário considerado nesse estudo é o de um VANT realizando captura de vídeo em tempo real, transmitindo-o a uma base terrestre por meio de rede sem fio. O problema consiste no fato de não ser possível garantir uma taxa de transmissão contínua, com banda estável. Isso ocorre devido a fatores como a velocidade da aeronave (da ordem centenas de km/h), irregularidades de terreno (impedindo a linha de visada do enlace de transmissão), ou do clima, como tempestades que podem interferir na transmissão da RF. Por fim, os movimentos que o VANT pode realizar no vôo (Rolagem, Arfagem ou Guinada) podem prejudicar a disponibilidade do link. Dessa forma, é necessário que seja realizada adaptação de vídeo de acordo com a banda disponível. Assim, quando a qualidade do enlace for degradada, deverá ser realizada uma redução no tamanho do vídeo, evitando a interrupção na transmissão. Por outro lado, a adaptação também deverá fazer com que a banda disponível seja utilizada, evitando o envio de vídeos com qualidade inferior à que seria possível para determinado valor de largura de banda. Nesse trabalho será considerada a faixa de valores de largura de banda de 8 Mbps até zero. Para realizar a adaptação será utilizado o padrão H.264/AVC com codificação escalável / Unmanned Aerial Vehicles (UAVs) are being increasingly used in several countries, both in the military and civilian areas. In this study we consider an UAV equipped with a camera, capturing video for a real-time transmission to a ground-base using wireless network. The problem is that its not possible to ensure a continuous transmission rate, with stable bandwidth. That occurs due to factors like the speed of the aircraft, irregularities of terrain, or the weather (as storms, heat and fog, for instance, can interfere with RF transmission). Finally, the movements that the UAV can perform in flight (Roll, pitch and yaw) can impair link availability. Thus, it is necessary to perform an adaptation of video according to the available bandwidth. When the link quality is degraded, a reduction in the resolution of the video must be performed , avoiding interruption of the transmission. Additionally, adaptation must also provide that all the available bandwidth is used, avoiding sending the video with lower quality that would be possible for a given value bandwidth. In this work we propose a system which can vary the total amount of data being transmitted, by adjusting the compression parameters of the video. We manage to produce a system which uses the range from 8 Mbps up to zero. We use the H.264/AVC Codec, with scalable video coding

Technology advancement increases the usage of the internet day by day. One of the most used internet services is video streaming. The major advantage of video streaming is that it allows long distance communication between people without any delay. It is known that streaming video is one of the fastest growing industries, and it has been very beneficial to the world. As the use of video streaming is increasing rapidly, it is essential to have security for video streaming. A lot of methods to secure video streaming came into existence like authentication, protocol, or some secure web hosting sites. Every method is to secure the transmission of video streaming, so these methods use either key or any other additional things to secure it. Our approach is to have a secure video streaming method without using any additional key or software. So, the method here is to encrypt the video directly and then encode it to be in the streaming format. The encryption of the video is done by a method called obscuring method. And the performance evaluation is done to the method so as to check the stability and feasibility of the system.

L’objectif de cette thèse est de trouver de nouvelles méthodes de compression sémantique compatible avec un encodeur classique tel que H.264/AVC. . L’objectif principal est de maintenir la sémantique et non pas la qualité globale. Un débit cible de 300 kb/s a été fixé pour des applications de sécurité et de défense Pour cela une chaine complète de compression a dû être réalisée. Une étude et des contributions sur les modèles de saillance spatio-temporel ont été réalisées avec pour objectif d’extraire l’information pertinente. Pour réduire le débit, une méthode de redimensionnement dénommée «seam carving » a été combinée à un encodeur H.264/AVC. En outre, une métrique combinant les points SIFT et le SSIM a été réalisée afin de mesurer la qualité des objets sans être perturbée par les zones de moindre contenant la majorité des artefacts. Une base de données pouvant être utilisée pour des modèles de saillance mais aussi pour de la compression est proposée avec des masques binaires. Les différentes approches ont été validées par divers tests. Une extension de ces travaux pour des applications de résumé vidéo est proposée. / The objective of this thesis is to find new methods for semantic video compatible with a traditional encoder like H.264/AVC. The main objective is to maintain the semantic and not the global quality. A target bitrate of 300 Kb/s has been fixed for defense and security applications. To do that, a complete chain of compression has been proposed. A study and new contributions on a spatio-temporal saliency model have been done to extract the important information in the scene. To reduce the bitrate, a resizing method named seam carving has been combined with the H.264/AVC encoder. Also, a metric combining SIFT points and SSIM has been created to measure the quality of objects without being disturbed by less important areas containing mostly artifacts. A database that can be used for testing the saliency model but also for video compression has been proposed, containing sequences with their manually extracted binary masks. All the different approaches have been thoroughly validated by different tests. An extension of this work on video summary application has also been proposed.

This study contributes to the domain of application specific adaptive hardware architectures with a design approach on processing element array, interconnect structure and memory interface concurrently. As summarized below, our architectural design choices push the limits of on-chip data reuse and avoid redundant computations that are essential for the high throughput, small area, and low power demands of the consumer market.Motion estimation (ME) is a key component in the H.264/AVC standard. Full Search (FS) based ME achieves optimal peak signal-to-noise-ratio (PSNR), and is the most adopted algorithm for developing hardware motion estimators. In this study, we first design a variable block size motion estimation (VBSME) engine based on hybrid grained processing elements (PEs) and a 2D programmable interconnect structure, which is adaptive to all block size configurations of H.264. PEs operate in bit-serial manner using MSB-first arithmetic for early termination to reduce the amount of computations, and the 2D architecture enables on-chip data reuse between neighboring PEs in a bit-by-bit pipelined fashion. Our design reduces the gate count by 7x compared to its ASIC counterpart, operates at a comparable frequency while sustaining 30 and 60 frames per second (fps); and outperforms bit parallel and bit serial architectures in terms of throughput and performance per gate.Numerous fast search algorithms (diamond, hexagon, three-step, etc.) have been developed to reduce the computation burden and the excessive amount of memory transactions required by FS, with a compromise in compression quality. We improve our VBSME engine and introduce the first adaptive ME architecture that provides the end user with the flexibility of choosing between the high quality video service during power-rich state (FS mode), and extended video service (fast search mode). We resolve the irregular indexing scheme challenge of three-step search (3SS) by introducing an on-chip buffer structure with a memory interface, which is adaptive to data access patterns of the FS and 3SS methods. The architecture sustains the real time CIF format (352x288) video encoding at 30fps with an operational frequency as low as 17.6MHz, and consumes 1.98mW based on the 45nm technology, outperforming all other FS and 3SS architectures.

L'étude des réseaux sur puces (NoC) est un domaine de recherche qui traite principalement la communication globale dans les systèmes sur puce (SoC). La topologie choisie et l'algorithme de routage jouent un rôle essentiel durant la phase de conception des architectures NoC. La modélisation des structures répétitives telles que les topologies des réseaux sur puce sous des formes graphiques pose un défi particulier. Cet aspect peut être rencontré dans les applications orienté contrôle/données intensif tel que le codeur vidéo H.264. Model Driven Engineering est une méthodologie de développement logiciel où le système complet est modélisé à un niveau d'abstraction élevé en utilisant un langage de modélisation unifié comme l’UML/MARTE. Le profil UML pour la modélisation et l'analyse des systèmes embarqués en temps réel (MARTE) est la norme actuelle pour la modélisation des SoCs.Cette thèse décrit une méthodologie adéquate pour la modélisation des NoCs en utilisant le profil MARTE. L'étude proposée a montré que le paquetage RSM (Repetitive Structure Modeling) du profil MARTE est assez puissant pour modéliser différent types de topologies. En utilisant cette méthodologie, plusieurs aspects tels que l’algorithme de routage sont modélisés en se basant sur les machines d'état. Ceci permet au profil MARTE à être assez complet pour la modélisation d'un grand nombre d’architectures de NoCs. Certains travaux sont en cours pour synthétiser ces réseaux, en VHDL à partir de ces modèles. Pour la validation de la méthodologie proposée, une approche de co-design a été étudiée par l’implémentation d'un système de codage vidéo H.264 sur un NoC de type Diagonal Mesh en utilisant model en « Y » de l’outil Gaspard2. Avant de passer à l'association de l'application/architecture, une optimisation architecturale ciblant la réduction de la puissance consommée du module le plus critique (Estimateur de Mouvement) de l'application a été effectué. Ainsi, une architecture VLSI flexible d’un estimateur de mouvement à blocks variables (FSVBSME) a été proposée. / The study of Networks on Chips (NoCs) is a research field that primarily addresses the global communication in Systems-on-Chip (SoCs). The selected topology and the routing algorithm play a prime role during the design of NoC architectures.The modeling of repetitive structures such as network on chip topologies in graphics forms poses a particular challenge. This aspect may be encountered in intensive data/control oriented applications such as H.264 video coder. Model driven engineering is a software development methodology where the complete system is modeled at a high abstraction level using a modeling language as UML/MARTE. The UML profile for Modeling and Analysis of Real-Time Embedded systems (MARTE) is the current standard for the SoCs modeling. This thesis describes an adequate methodology for modeling NoCs by using the MARTE standard profile. The proposed study has shown that the Repetitive Structure Modeling (RSM) package of MARTE profile is powerful enough for modeling different topologies. By using this methodology, several aspects such as routing algorithm are modeled based finite state machines. This allows to the MARTE profile to be complete enough for modeling a large number of NoCs architectures. Some work is on-going to synthesize such networks in VHDL from such models. While validating the proposed methodology, a co-design approach has been studied by mapping a H264 video coding system onto a Diagonal Mesh NoC by using the Y Chart of Gaspard2 tool. Before allowing the association of the application/architecture, an architectural optimization targeting power minimization of the most critical module of the application has been performed. So, a flexible VLSI architecture for full-search VBSME (FSVBSME) has been proposed.

<p>Communication over today’s IP-based networks are to some extent subject to packet loss. Most real-time applications, such as video streaming, need methods to hide this effect, since resending lost packets may introduce unacceptable delays. For IP-based video streaming applications such a method is referred to as a <em>packet loss concealment </em>scheme.</p><p>In this thesis a recently proposed mixture model and least squares-based packet loss concealment scheme is implemented and evaluated together with three more well known concealment methods. The JM reference software is used as basis for the implementation, which is a public available software codec for the H.264 video coding standard. The evaluation is carried out by comparing the schemes in terms of objective measurements, subjective observations and a study with human observers.</p><p>The recently proposed packet loss concealment scheme shows good performance with respect to the objective measures, and careful observations indicate better concealment of scenes with fast motion and rapidly changing video content. The study with human observers verifies the results for the case when a more sophisticated packetization technique is used.</p><p>A new packet loss concealment scheme, based on joint modeling of motion vectors and pixels, is also investigated in the last chapter as an additional contribution of the thesis.</p>

We develop a framework that uses visual attention analysis combined with temporal coherence to detect the attended region from a H.264 video bitstream, and display it on a small screen. A visual attention module based upon Walther and Koch's model gives us the attended region in I-frames. We propose a temporal coherence matching framework that uses the motion information in P-frames to extend the attended region over the H.264 video sequence. Evaluations show encouraging results with over 80% successful detection rate for objects of interest, and 85% respondents claiming satisfactory output.

We develop a framework that uses visual attention analysis combined with temporal coherence to detect the attended region from a H.264 video bitstream, and display it on a small screen. A visual attention module based upon Walther and Koch's model gives us the attended region in I-frames. We propose a temporal coherence matching framework that uses the motion information in P-frames to extend the attended region over the H.264 video sequence. Evaluations show encouraging results with over 80% successful detection rate for objects of interest, and 85% respondents claiming satisfactory output.

H.264/MPEG-4 part 10 or Advanced Video Coding (AVC) is a standard for video compression. MPEG-4 is currently one of the most widely used formats for recording, compression and distribution of high definition video. One feature of the AVC codec is the inclusion of an in-loop deblocking filter. The goal of the deblocking filter is to remove blocking artifacts that exist at macroblock boundaries. However, due to the complexity of the deblocking algorithm, the filter can easily account for one-third of the computational complexity of a decoder. In this thesis, a modification to the deblocking algorithm given in the AVC standard is presented. This modification allows the algorithm to finish the filtering of a macroblock to finish twenty clock cycles faster than previous single filter designs. This thesis also presents a hardware architecture of the H.264 deblocking filter to be used in the H.264 decoder. The developed architecture allows the filtering of videos streams using 4:2:2 chroma subsampling and 10-bit pixel precision in real-time. The filter was described in VHDL and synthesized for a Spartan-6 FPGA device. Timing analysis showed that is was capable of filtering a macroblock using 4:2:0 chroma subsampling in 124 clock cycles and 4:2:2 chroma subsampling streams in 162 clock cycles. The filter can also provide real-time deblocking of HDTV video (1920x1080) of up to 988 frames per second.

This thesis proposes a strategy for enhancing the efficiency of scalable H.264/AVC video transmission over multi-input multi-output (MIMO) and cooperative communication systems. For scalable video coding (SVC) transmission over MIMO wireless systems, a channel selection algorithm is used to enhance transmission rate. The proposed algorithm allows SVC layers to select channels individually in wireless MIMO systems based on channel state information for transmission rate enhancement. Here, this difficult problem is converted into a mathematical optimization problem to improve SVC performance during video transmission. Experimental results show that the proposed method achieves a higher transmission rate over MIMO systems compared to the existing scheme. For SVC transmission over cooperative communication systems, the algorithm allows each SVC layer to choose an appropriate relay based on channel conditions and SVC layer priority. Thus, SVC data is protected effectively. Experimental results show that video quality obtained by the algorithm exceeds that of non-cooperative systems.