RecycleNet: Intelligent Waste Sorting Using Deep Neural Networks

2018-07-05
Bircanoglu, Cenk
Atay, Meltem
Beser, Fuat
Genc, Ozgun
Kizrak, Merve Ayyuce
Waste management and recycling is the fundamental part a sustainable economy. For more efficient and safe recycling, it is necessary to use intelligent systems instead of employing humans as workers in the dump-yards. This is one of the early works demonstrating the efficiency of latest intelligent approaches. In order to provide the most efficient approach, we experimented on well-known deep convolutional neural network architectures. For training without any pre-trained weights, Inception-Resnet, Inception-v4 outperformed all others with 90% test accuracy. For transfer learning and fine-tuning of weight parameters using ImageNet, DenseNet121 gave the best result with 95% test accuracy. One disadvantage of these networks, however, is that they are slightly slower in prediction time. To enhance the prediction performance of the models we altered the connection patterns of the skip connections inside dense blocks. Our model RecycleNet is carefully optimized deep convolutional neural network architecture for classification of selected recyclable object classes. This novel model reduced the number of parameters in a 121 layered network from 7 million to about 3 million.
IEEE (SMC) International Conference on Innovations in Intelligent Systems and Applications (INISTA)

Suggestions

Transportation and recycling effects on the household textile waste of Eskisehir using LCA
Sipahi, İlayda; Kurt, Zöhre; Department of Environmental Engineering (2021-8-16)
Textile is a material that could be diversified the most, therefore considerations of textile waste recycling/ reuse and the effects of its landfill should be carefully considered. In this thesis, transportation and recycling effects on the household textile waste of Eskişehir was investigated by using Life Cycle Assessment (LCA). Disregarding consumer using stage, the entire life cycle of waste produced from cotton was evaluated. LCA inventory was included for occupation of land, water consumption, c...
Solvent recovery from photolithography wastes using cellulose ultrafiltration membranes
Savaş, Aygen; Çulfaz Emecen, Pınar Zeynep (2022-04-05)
Solvent recycling and reuse are indispensable for ensuring a sustainable chemical industry and circular economy. In this study we report the fabrication of cellulose ultrafiltration membranes and their application in recovery of propylene glycol methyl ether acetate (PGMEA) used as developer solvent in SU-8 photolithography. Cellulose membranes were fabricated via alkaline hydrolysis of cellulose acetate membranes in aqueous NaOH. Membrane permeance and molecular weight cut-off (MWCO) were tuned via changin...
Energy performance of smart buildings: simulating the impact of active systems and passive strategies
Tetik, Buğra; Elias Özkan, Soofia Tahira; Department of Building Science in Architecture (2014)
Energy efficiency is one of the most important attempts in the world because of various environmental, economical and developmental aspects of energy. In this context, energy performance of buildings has been a critical issue since buildings constitute approximately half of total energy consumption. The concept of smart building which has been attractive recently, contributes to the issue with smart technologies; while some passive design techniques which have been used throughout the history are still appl...
Energy performance of duble-skin fa̧acades in intelligent office buildings : a case study in Germany
Bayram, Ayça; Elias Özkan, Soofia Tahira; Department of Architecture (2003)
The building industry makes up a considerable fraction of world̕s energy consumption. The adverse effects of a growing energy demand such as depletion in fossil fuel reserves and natural resources hassled the building industry to a search for new technologies that result in less energy consumption together with the maximum utilization of natural resources. Energy- and ecology-conscious European countries incorporated the well-being of occupants while conducting research on innovative technologies. In view o...
Spreparation of multifunctional materials for photocatalytic applications
Uzun, Cere; Volkan, Mürvet; Department of Chemistry (2019)
Due to the increasingly polluted environment and the limited energy reserves, the development of high efficient renewable technologies, green energy sources and ecofriendly methods for environmental remediation and energy production is highly important. Hydrogen (H2), as a clean and carbonless energy source, is of great potential in solving the environmental pollution and energy shortage. Turkey is a country that clued-in textile production. But the widespread discharge of wastewaters from the textile indus...
Citation Formats
C. Bircanoglu, M. Atay, F. Beser, O. Genc, and M. A. Kizrak, “RecycleNet: Intelligent Waste Sorting Using Deep Neural Networks,” Thessaloniki, Greece, 2018, p. 0, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/67997.