EUVIMEDEUVIMED entdecken
Crossref

SEVEN CLASS SOLID WASTE MANAGEMENT-HYBRID FEATURES BASED DEEP NEURAL NETWORK

Krishna Bikram Shah, S. Visalakshi, Ranjit Panigrahi

International Journal of System Design and Computing · 28.09.2023 · 01

DOI: 10.66135/ijsdc010101 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of System Design and Computing, online erschienen am 28.09.2023. Verfasst von Krishna Bikram Shah und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To keep cities clean, municipal agencies, non-profit organizations, and the private sector all provide solid management as one of their core essential services. Waste management includes the gathering, moving, handling, and getting rid of solid waste as well as monitoring and control. Due to urbanization, India is experiencing a significant environmental problem with solid waste management. One of cities' largest issues is solid garbage. The proposed techniques are employed in this work to categories the seven-class solid waste into biodegradable and non-biodegradable waste. Images from the training and testing datasets are used for each input dataset. Feature extraction is the next step, where MeQryEP and PCA (Principal Component Analysis) are employed. It can be mentioned as hybrid features (HF). HF are utilized to extract the texture and shape of the provided images, respectively. The extracted images are then combined. The collected seven class solid waste image is then classified under biodegradable and non-biodegradable garbage using DNN. The experimental results demonstrated that classification accuracy using DNN is 99.16%. Additionally, after a few epoch calculations, deep neural network demonstrated great accuracy, nearly 100%. The proposed seven class solid waste management improves the overall accuracy 9.23%, 0.24%, 37.47%, 3.59% better than PCA+MFCC, PCA+NFNN, PCA+SIFT, PCA+SVM.

Zitieren

APA

Shah, K. B., Visalakshi, S., & Panigrahi, R. (2023). SEVEN CLASS SOLID WASTE MANAGEMENT-HYBRID FEATURES BASED DEEP NEURAL NETWORK. International Journal of System Design and Computing, 01. https://doi.org/10.66135/ijsdc010101

Vancouver

Shah KB, Visalakshi S, Panigrahi R. SEVEN CLASS SOLID WASTE MANAGEMENT-HYBRID FEATURES BASED DEEP NEURAL NETWORK. International Journal of System Design and Computing. 2023:01. doi:10.66135/ijsdc010101

BibTeX
@article{Shah2023,
  author = {Shah, Krishna Bikram and Visalakshi, S. and Panigrahi, Ranjit},
  title = {SEVEN CLASS SOLID WASTE MANAGEMENT-HYBRID FEATURES BASED DEEP NEURAL NETWORK},
  journal = {International Journal of System Design and Computing},
  year = {2023},
  pages = {01},
  doi = {10.66135/ijsdc010101}
}
Kostenlos starten →