Weed detection using image processing under different illumination for site-specific areas spraying

Agriculture, although known as the backbone of the Indian economy, is facing crisisin terms of production. One of the major issues in the agriculture sector is the growth of weeds among the crops. They compete with the desired plants for various resources and hence their growth must be inhibited. At present weeds are removed either manually, which is a time consuming and labour intensive task, or herbicides are being sprayed uniformly all over the field to keep them under check. Spraying of herbicide is very inefficient as the chemical contributes less to weed control and cause contamination of the environment. The main objective of this work is a weed control system that differentiates the weed from crops and restricts weed growth alone by the precise removal of the weed. This is implemented by capturing the images of the field at regular intervals and processing them with a Raspberry Pi board by making use of an image processing algorithm to differentiate the desired plants from the weeds. This is based on various features like colour and size of the crop and weed. Once the weeds are identified and located correctly through image processing, a signal is transmitted from the Raspberry Pi board to turn on the weed cutting system. The selective activation of the weed removal system helps in the precise removal of the weeds and this provides a better environment for the desired plants to grow well.

Download Full-text

OpenWeedLocator (OWL): an open-source, low-cost device for fallow weed detection

Scientific Reports ◽

10.1038/s41598-021-03858-9 ◽

2022 ◽

Vol 12 (1) ◽

Author(s):

Guy Coleman ◽

William Salter ◽

Michael Walsh

Keyword(s):

Open Source ◽

Weed Control ◽

Technology Development ◽

Low Cost ◽

Yield Potential ◽

Low Density ◽

Weed Detection ◽

Site Specific ◽

South Wales ◽

Control Community

AbstractThe use of a fallow phase is an important tool for maximizing crop yield potential in moisture limited agricultural environments, with a focus on removing weeds to optimize fallow efficiency. Repeated whole field herbicide treatments to control low-density weed populations is expensive and wasteful. Site-specific herbicide applications to low-density fallow weed populations is currently facilitated by proprietary, sensor-based spray booms. The use of image analysis for fallow weed detection is an opportunity to develop a system with potential for in-crop weed recognition. Here we present OpenWeedLocator (OWL), an open-source, low-cost and image-based device for fallow weed detection that improves accessibility to this technology for the weed control community. A comprehensive GitHub repository was developed, promoting community engagement with site-specific weed control methods. Validation of OWL as a low-cost tool was achieved using four, existing colour-based algorithms over seven fallow fields in New South Wales, Australia. The four algorithms were similarly effective in detecting weeds with average precision of 79% and recall of 52%. In individual transects up to 92% precision and 74% recall indicate the performance potential of OWL in fallow fields. OWL represents an opportunity to redefine the approach to weed detection by enabling community-driven technology development in agriculture.

Download Full-text

Weed Detection in Farm Crops using Parallel Image Processing

2018 Conference on Information and Communication Technology (CICT) ◽

10.1109/infocomtech.2018.8722369 ◽

2018 ◽

Cited By ~ 1

Author(s):

S Umamaheswari ◽

R Arjun ◽

D Meganathan

Keyword(s):

Image Processing ◽

Weed Detection ◽

Parallel Image Processing ◽

Parallel Image

Download Full-text

Categorization of Plant Sapling using Deep Learning

International Journal of Engineering and Advanced Technology - Regular Issue ◽

10.35940/ijeat.e7569.088619 ◽

2019 ◽

Vol 8 (6) ◽

pp. 289-292

Keyword(s):

Neural Networks ◽

Image Processing ◽

Deep Learning ◽

Food Production ◽

Limited Resources ◽

Row Spacing ◽

Weed Detection ◽

Data Representations ◽

Rgb Images ◽

Plant Seedlings

According to the latest research, the current increment of the food production will not be able to satisfy the market due to the lack of farmers, and land areas with limited resources. Weeds are grown along with the plant seedlings, The Amount of water and fertilizers needed for the plant seedlings, Crop placement, and row spacing, are being the Major problems. The usage of deep learning using digital image processing could be an efficient way to overcome these problems. Deep Learning is based on Data Representations, Artificial and Neural networks. Plant species can be recognized using RGB images especially in the problem of weed detection. The resources and the row spacing needed for the seedlings can be fulfilled by recognizing the image with the preloaded datasets of the seedling

Download Full-text

Weed Mapping with UAS Imagery and a Bag of Visual Words Based Image Classifier

Remote Sensing ◽

10.3390/rs10101530 ◽

2018 ◽

Vol 10 (10) ◽

pp. 1530 ◽

Cited By ~ 11

Author(s):

Michael Pflanz ◽

Henning Nordmeyer ◽

Michael Schirrmann

Keyword(s):

Plant Protection ◽

Unmanned Aircraft ◽

Aerial Images ◽

Support Vector ◽

Bag Of Visual Words ◽

Weed Species ◽

Weed Detection ◽

Site Specific ◽

Wheat Field ◽

Visual Words

Weed detection with aerial images is a great challenge to generate field maps for site-specific plant protection application. The requirements might be met with low altitude flights of unmanned aerial vehicles (UAV), to provide adequate ground resolutions for differentiating even single weeds accurately. The following study proposed and tested an image classifier based on a Bag of Visual Words (BoVW) framework for mapping weed species, using a small unmanned aircraft system (UAS) with a commercial camera on board, at low flying altitudes. The image classifier was trained with support vector machines after building a visual dictionary of local features from many collected UAS images. A window-based processing of the models was used for mapping the weed occurrences in the UAS imagery. The UAS flight campaign was carried out over a weed infested wheat field, and images were acquired between a 1 and 6 m flight altitude. From the UAS images, 25,452 weed plants were annotated on species level, along with wheat and soil as background classes for training and validation of the models. The results showed that the BoVW model allowed the discrimination of single plants with high accuracy for Matricaria recutita L. (88.60%), Papaver rhoeas L. (89.08%), Viola arvensis M. (87.93%), and winter wheat (94.09%), within the generated maps. Regarding site specific weed control, the classified UAS images would enable the selection of the right herbicide based on the distribution of the predicted weed species.

Download Full-text