An Extensive Survey of Deep learning-based Crop Yield Prediction Models for Precision Agriculture

EasyChair Preprint 8374

Abstract

Precision agriculture, as the trademark of the agriculture 4.0 period, has assured to reform agricultural practices using monitoring and intervention technologies to increase productivity and decrease the environmental impact. Computer vision (CV) and deep learning (DL) models are commonly used as key enablers for precision agriculture. CV technologies utilize digital images for the interpretation and understanding of the world to offer precise, region orient details about the crops and respective surroundings. Today, CV has been widely employed to support precision agriculture processes like crop yield prediction (CYP), crop monitoring, weed control, plant disease detection, weed detection, etc. CYP is a significant process for decision making at the national and regional levels. Several machine learning (ML) and DL based models have been presented for accurate CYP. Therefore, this paper reviews existing DL-based CYP models developed for precision agriculture. In this view, the major aim of the review is to identify, group, and discuss the existing intelligent agriculture approaches. The existing methods are surveyed based on the underlying techniques, objectives, dataset used, and available datasets. The outcome of the survey pointed out the significance of applying DL models for CYP in precision agriculture.

Keyphrases: Crop yield prediction, Precision Agriculture, computer vision, deep learning, image processing

@booklet{EasyChair:8374,
  author    = {Srilatha Toomula and Sudha Pelluri},
  title     = {An Extensive Survey of Deep learning-based Crop Yield Prediction Models for Precision Agriculture},
  howpublished = {EasyChair Preprint 8374},
  year      = {EasyChair, 2022}}