EUVIMEDEUVIMED entdecken
Crossref

Pest Classification using Morphological Processing in Deep Learning

Sanjyot Thuse, Meena Chavan

International Journal of Electronics and Computer Applications · 2024

DOI: 10.70968/ijeaca.v1i1.thuse ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Electronics and Computer Applications, 2024. Verfasst von Sanjyot Thuse und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Agriculture relies heavily on the prompt detection of pests. There are numerous technologies for identifying pests, but almost all of them are susceptible to misclassification due to inadequate lighting, background distractions, a diversity of collection techniques. Thus, pests that are only partially visible or oriented differently. This misclassification could result in a significant yield loss. We presented an architecture that would use skeletonization together with neural networks as classifiers to give excellent classification accuracy under the aforementioned parameters in order to alleviate this problem. The paper compares the performance of CNN. CNN with VGG16 and the proposed system on accuracy metric. The results obtained are on disoriented images. From the findings it is observed that for data augmented dataset CNN gives 80 accuracy, CNN with VGG16 gives 95 accuracy, and CNN with feature extractor, skeletonization achieved good accuracy which is up to 98. A web app and an android app is also developed to classify the pest which will help the farmers to identify the name of the pest without going into technical details. This framework will surely help farmers in identifying pest names instantly which will later help in identifying the name and quantity of pesticide.

Zitieren

APA

Thuse, S., & Chavan, M. (2024). Pest Classification using Morphological Processing in Deep Learning. International Journal of Electronics and Computer Applications. https://doi.org/10.70968/ijeaca.v1i1.thuse

Vancouver

Thuse S, Chavan M. Pest Classification using Morphological Processing in Deep Learning. International Journal of Electronics and Computer Applications. 2024. doi:10.70968/ijeaca.v1i1.thuse

BibTeX
@article{Thuse2024,
  author = {Thuse, Sanjyot and Chavan, Meena},
  title = {Pest Classification using Morphological Processing in Deep Learning},
  journal = {International Journal of Electronics and Computer Applications},
  year = {2024},
  doi = {10.70968/ijeaca.v1i1.thuse}
}
Kostenlos starten →