Review on Tomato Ripe Detection and Segmentation Using Deep learning Models for Sustainable Agricultural Development

Abstract

Using natural resources to maximize yields is possible when .precision agriculture is used in a diversified environment. Automating agriculture can reduce resource consumption and enhance food quality. Sowing, monitoring, controlling weeds, managing pests, and harvesting crops are all possible with agricultural robots. To estimate crop production, it is necessary to physically count fruits, flowers, or fruits at various stages of growth. Precision and dependability are provided by remote sensing technologies for agricultural production forecasting and estimation. Automated image analysis using deep learning and computer vision (CV) produces exact field maps. In this review, deep learning (DL) techniques were found to improve the accuracy of smart farming, so we present different methodologies to automate the detection of agricultural yields using virtual analysis and classifiers. The smart farming will generate a sustainable agricultural development.