Classificazione di oggetti in immagini attraverso il modello Bag of Visual Words

Generic object recognition is an important function of the human visual system and everybody finds it highly useful in their everyday life. For an artificial vision system it is a really hard, complex and challenging task because instances of the same object category can generate very different images, depending of different variables such as illumination conditions, the pose of an object, the viewpoint of the camera, partial occlusions, and unrelated background clutter. The purpose of this thesis is to develop a system that is able to classify objects in 2D images based on the context, and identify to which category the object belongs to. Given an image, the system can classify it and decide the correct categorie of the object. Furthermore the objective of this thesis is also to test the performance and the precision of different supervised Machine Learning algorithms in this specific task of object image categorization. Through different experiments the implemented application reveals good categorization performances despite the difficulty of the problem. However this project is open to future improvement; it is possible to implement new algorithms that has not been invented yet or using other techniques to extract features to make the system more reliable. This application can be installed inside an embedded system and after trained (performed outside the system), so it can become able to classify objects in a real-time. The information given from a 3D stereocamera, developed inside the department of Computer Engineering of the University of Bologna, can be used to improve the accuracy of the classification task. The idea is to segment a single object in a scene using the depth given from a stereocamera and in this way make the classification more accurate.

Contributo sperimentale alla ridefinizione della diffusione in parete della camera di prova della galleria del vento Dallara

Joseph Nicolas Cugnot built the first primitive car in 1769 and approximately one hundred year later the first automotive race took place. Thanks to this, for the first time the aerodynamics principles began to be applied to cars. The aerodynamic study of a car is important to improve the performance on the road, or on the track. It purposely enhances the stability in the turns and increases the maximum velocity. However, it is also useful, decrease the fuel consumption, in order to reduce the pollution. Given that cars are a very complex body, the aerodynamic study cannot be conducted following an analytical method, but it is possible, in general, to choose between two different approaches: the numerical or the experimental one. The results of numerical studies depend on the computers’ potential and on the method use to implement the mathematical model. Today, the best way to perform an aerodynamic study is still experimental, which means that in the first phase of the design process the study is performed in a wind tunnel and in later phases directly on track. The automotive wind tunnels are singular mainly due to the test chamber, which typically contains a ground simulation system. The test chamber can have different types of walls: open walls, closed walls, adaptive walls or slotted walls. The best solution is to use the slotted walls because they minimize the interference between the walls and the streamlines, the interaction between the flow and the environment, and also to contain the overall costs. Furthermore, is necessary minimize the boundary layer at the walls, without accelerating the flow, in order to provide the maximum section of homogeneous flow. This thesis aims at redefining the divergent angle of the Dallara Automobili S.P.A. wind tunnel’s walls, in order to improve the overall homogeneity. To perform this study it was necessary to acquire the pressure data of the boundary layer, than it was created the profile of the boundary layer velocity and, to minimize the experimental errors, it was calculated the displacement thickness. The results obtained shows, even if the instrument used to the experiment was not the best one, that the boundary layer thickness could be minor in case of a low diffusion angle. So it is convenient to perform another experiment with a most sensitive instrument to verified what is the better wall configuration.