No description available.

Thesis (M. Sc.)--University of Hong Kong, 1999. / Includes bibliographical references (leaves 45-47).

Univ., Diss.--Göteborg, 2007. / Enth. 4 Beitr.

<p>  </p> <p>The proper classification and disposal of waste are crucial in reducing environmental impacts and promoting sustainability. Several solid waste classification systems have been developed over the years, ranging from manual sorting to mechanical and automated sorting. Manual sorting is the oldest and most commonly used method, but it is time-consuming and labor-intensive. Mechanical sorting is a more efficient and cost-effective method, but it is not always accurate, and it requires constant maintenance. Automated sorting systems use different types of sensors and algorithms to classify waste, making them more accurate and efficient than manual and mechanical sorting systems. In this thesis, we propose the development of an intelligent solid waste detection, classification and tracking system using artificial deep learning techniques. To address the limited samples in the TrashNetV2 dataset and enhance model performance, a data augmentation process was implemented. This process aimed to prevent overfitting and mitigate data scarcity issues while improving the model's robustness. Various augmentation techniques were employed, including random rotation within a range of -20° to 20° to account for different orientations of the recycled materials. A random blur effect of up to 1.5 pixels was used to simulate slight variations in image quality that can arise during image acquisition. Horizontal and vertical flipping of images were applied randomly to accommodate potential variations in the appearance of recycled materials based on their orientation within the image. Additionally, the images were randomly scaled to 416 by 416 pixels, maintaining a consistent image size while increasing the dataset's overall size. Further variability was introduced through random cropping, with a minimum zoom level of 0% and a maximum zoom level of 25%. Lastly, hue variations within a range of -20° to 20° were randomly introduced to replicate lighting condition variations that may occur during image acquisition. These augmentation techniques collectively aimed to improve the dataset's diversity and the model's performance. In this study, YOLOv8, EfficientNet-B0 and VGG16 architectures were evaluated, and stochastic gradient descent (SGD) and Adam were used as the optimizer. Although, SGD provided better test accuracies compared to Adam. </p> <p>Among the three models, YOLOv8 showed the best performance, with the highest average precision mAP of 96.5%. YOLOv8 emerges as the top performer, with ROC values varying from 92.70% (Metal) to 98.40% (Cardboard). Therefore, the YOLOv8 model outperforms both VGG16 and EfficientNet in terms of ROC values and mAP. The findings demonstrate that our novel classifier tracker system made of YOLOv8, and supervision algorithms surpass conventional deep learning methods in terms of precision, resilience, and generalization ability. Our contribution to waste management is in the development and implementation of an intelligent solid waste detection, classification, and tracking system using computer vision and deep learning techniques. By utilizing computer vision and deep learning algorithms, our system can accurately detect, classify, and localize various types of solid waste on a moving conveyor, including cardboard, glass, metal, paper, and plastic. This can significantly improve the efficiency and accuracy of waste sorting processes.</p> <p>This research provides a promising solution for detection, classification, localization, and tracking of solid waste materials in real time system, which can be further integrated into existing waste management systems. Through comprehensive experimentation and analysis, we demonstrate the superiority of our approach over traditional methods, with higher accuracy and faster processing times. Our findings provide a compelling case for the implementation of intelligent solid waste sorting.</p>

Au cours de ces dernières années, les Déchets Solides Municipaux (DSM) se sont révélés comme une problématique environnementale et économique majeure dans tous les pays. Les quantités collectées et les dépenses publiques nécessaires à leurs gestions croissent de façon insoutenable et ce, particulièrement dans les pays en développement (y compris ceux en transition). Face à ce constat, notre thèse a pour objectif de proposer des éléments de compréhension ainsi que des recommandations pour les politiques publiques. Pour cela, nous avons combiné une approche macroéconomique, pour étudier les liens entre la quantité de DSM et la croissance économique, à une approche microéconomique centrée sur la question du financement par une taxe incitative. Les contributions de cette thèse sont de trois ordres et ont pour référence les pays méditerranéens. La première contribution a consisté à tester la viabilité de l’hypothèse de la Courbe Environnementale de Kuznets (CEK) dans un contexte d’ouverture au commerce international. Cette première approche a permis de déceler une relation monotone croissante entre l’intensité de la production des DSM et la croissance économique sur la période 1990-2010 et ce, quel que soit le niveau de revenu des pays. Ce qui nous conduit à la conclusion d’une incompatibilité entre les objectifs de croissance et de prévention de la production des DSM. La deuxième contribution a été l’occasion de considérer, dans un modèle théorique, le recyclage informel qui est une caractéristique commune au pays en développement. L’objectif de cette contribution a été d’une part d’analyser l’impact du recyclage informel sur l’efficacité d’une politique de taxation incitative et d’autre part, de déterminer les conditions optimales d’une telle politique en présence du recyclage informel. En étudiant l’efficacité d’une politique de taxation directe de type tarification à l’acte et d’une politique de taxation indirecte de type Deposit and Refund System (DRS), nous montrons que la présence du recyclage informel ne permet pas de faire coïncider optimum social et équilibre du marché. Cependant, la politique DRS peut être optimale, à condition de subventionner à la fois le recyclage formel et informel. Enfin, la troisième contribution est une étude économétrique des impacts du système de taxation incitative d’un pays riche méditerranéen. Cette étude vient du constat selon lequel les pays riches, contrairement aux pays en développement, mettent en œuvre plusieurs taxes incitatives de façon concomitante. Nous évaluons l’efficacité du système de taxation de la France, qui avec ses trois taxes incitatives est un cas d’école en la matière. Nous proposons dans cette contribution, à l’aide de tests économétriques sur données départementales, une mesure de l’élasticité de la quantité de DSM collectés, valorisés et éliminés par rapport à la Redevance sur l’Enlèvement des Ordures Ménagères (REOM), la Responsabilité Élargie du Producteur (REP) et la Taxe Générales sur les Activités Polluantes (TGAP). Les résultats montrent une complémentarité des trois taxes avec une supériorité de la REOM pour inciter les ménages à la prévention et à la valorisation, et une supériorité de la REP pour inciter les collectivités locales à la substitution des technologies d’élimination à celles de valorisation / In last decades, Municipal Solid Waste (MSW) has become a major environmental and economic problem in many countries. The quantity of MSW collected and the expenditures necessary for its management have rapidly increased, particularly in developing countries (including those in transition). Our thesis aims to shed light on the relationship between MSW collection and economic growth and to propose how public policy can sustainably manage this pollution. To attempt our objective, we combined a macro and micro economic approaches in theoretical and empirical studies. The contributions of this thesis are threefold and have the scope of Mediterranean countries. First, we complete the empirical literature on the validation of the Environmental Kuznets Curve hypothesis (EKC) by studying the relationship between MSW collected and the economic growth over the period 1990-2010. The main added value of this study is the use of multiple imputations methodology to control for the sample bias due to missing values. We find out that MSW collected monotonically increase with income whatever countries’ income level. This leads us to conclude that the economic growth and MSW prevention are not compatible. So in a second contribution, we investigate, in a theoretical model, the efficiency of using market-based incentives to reach to objective of waste prevention and diversion from disposal in developing countries. We consider the common phenomenon of waste picking in these countries and we analyze the conditions in which a policy of Deposit and Refund System (DRS) can help to achieve the first best optimum. We first analyze the impact of waste picking on the effectiveness of the user fee and the DRS, then we show that social optimum and market equilibrium can be hold by taxing consumer goods and by subsidizing both formal recycling and waste picking. Our third contribution in this thesis is an econometric evaluation of French MSW taxing system in order to test the hypothesis of the complementarity of direct and indirect MSW incentive taxes in developed countries, which implement simultaneously several incentives taxes. Using data aggregate at French administrative departments level, we assess the prevention and substitution effects of the three incentive taxes namely the French user fee (La Redevance d’Enlèvement des Ordures Ménagères), the Extended Producer responsibility and disposal tax levied at landfill and incineration (la Taxe Générale sur les Activités Polluantes). We confirm the complementarity hypothesis of these taxes.