Efficient segmentation and classification of weed images using deep learning techniques
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Crop production is the main activity of agriculture, which is
newlineresponsible for global food security, nutrition and management. Crops are
newlinebeneficial plants that are mass-produced and utilized in the commercial process.
newlineIt is essential to look at the new trends, scientific approaches, and accelerators
newlinethat can support the agricultural activities. The farmer can reduce their workload
newlineby employing technology that improves the crop quality. Weeds are plant
newlinespecies that outcompete crops in agricultural fields, reducing crop production
newlineand causing severe economic harm to farmers. They increase farming costs and
newlineslow irrigation operations by making farm machinery difficult to operate.
newlineThe weed destruction process is necessary to save soil nutrients for crops, but
newlinethere are less harmful weed species that could be left alive to increase diversity
newlineon the farm. Computer vision technology has recently been used to do numerous
newlinesmart farming tasks, such as plant species type detection, plant disease
newlineclassification, and crop yield prediction.
newlineControlling weeds is a crucial part in raising agricultural output.
newlineIdentification of weeds and automation of weed removal are two crucial
newlineprocesses. However, locating and classifying the weeds in precise level is
newlineessential for effective weed removal process. A quick and accurate
newlineimplementation of plant detection and classification is required, that categorises
newlinethe vegetation into crops, weeds and non-harmful weeds in real-time. The weed
newlinemanagement process is intended to be fulfilled by any computer vision
newlinetechnologies such as deep learning, and autonomous robot.
newlineThe major objective of this research work is to segment and classify
newlinethe leaf images using deep learning techniques
newline