Dataset Information

A novel method for detecting morphologically similar crops and weeds based on the combination of contour masks and filtered Local Binary Pattern operators.

ABSTRACT: BACKGROUND:Weeds are a major cause of low agricultural productivity. Some weeds have morphological features similar to crops, making them difficult to discriminate. RESULTS:We propose a novel method using a combination of filtered features extracted by combined Local Binary Pattern operators and features extracted by plant-leaf contour masks to improve the discrimination rate between broadleaf plants. Opening and closing morphological operators were applied to filter noise in plant images. The images at 4 stages of growth were collected using a testbed system. Mask-based local binary pattern features were combined with filtered features and a coefficient k. The classification of crops and weeds was achieved using support vector machine with radial basis function kernel. By investigating optimal parameters, this method reached a classification accuracy of 98.63% with 4 classes in the "bccr-segset" dataset published online in comparison with an accuracy of 91.85% attained by a previously reported method. CONCLUSIONS:The proposed method enhances the identification of crops and weeds with similar appearance and demonstrates its capabilities in real-time weed detection.

SUBMITTER: Le VNT

PROVIDER: S-EPMC7055473 | biostudies-literature | 2020 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A novel method for detecting morphologically similar crops and weeds based on the combination of contour masks and filtered Local Binary Pattern operators.

Le Vi Nguyen Thanh VNT Ahderom Selam S Apopei Beniamin B Alameh Kamal K

GigaScience 20200301 3

<h4>Background</h4>Weeds are a major cause of low agricultural productivity. Some weeds have morphological features similar to crops, making them difficult to discriminate.<h4>Results</h4>We propose a novel method using a combination of filtered features extracted by combined Local Binary Pattern operators and features extracted by plant-leaf contour masks to improve the discrimination rate between broadleaf plants. Opening and closing morphological operators were applied to filter noise in plan ...[more]

PMID: 32129847

Similar Datasets

Project description:BackgroundThe outcome of the arms race between hosts and pathogens depends heavily on the interactions between their genetic diversity, population size and transmission ability. Theory predicts that genetically diverse hosts will select for higher virulence and more diverse pathogens than hosts with low genetic diversity. Cultivated hosts typically have lower genetic diversity and thus small effective population sizes, but can potentially harbour large pathogen population sizes. On the other hand, hosts, such as weeds, which are genetically more diverse and thus have larger effective population sizes, usually harbour smaller pathogen population sizes. Large pathogen population sizes may lead to more opportunities for mutation and hence more diverse pathogens. Here we test the predictions that pathogen neutral genetic diversity will increase with large pathogen population sizes and host diversity, whereas diversity under selection will increase with host diversity. We assessed and compared the diversity of a fungal pathogen, Rhynchosporium commune, on weedy barley grass (which have a large effective population size) and cultivated barley (low genetic diversity) using microsatellites, effector locus nip1 diversity and pathogen aggressiveness in order to assess the importance of weeds in the evolution of the neutral and selected diversity of pathogens.ResultsThe findings indicated that the large barley acreage and low host diversity maintains higher pathogen neutral genetic diversity and lower linkage disequilibrium, while the weed maintains more pathotypes and higher virulence diversity at nip1. Strong evidence for more pathogen migration from barley grass to barley suggests transmission of virulence from barley grass to barley is common.ConclusionsPathogen census population size is a better predictor for neutral genetic diversity than host diversity. Despite maintaining a smaller pathogen census population size, barley grass acts as an important ancillary host to R. commune, harbouring highly virulent pathogen types capable of transmission to barley. Management of disease on crops must therefore include management of weedy ancillary hosts, which may harbour disproportionate supplies of virulent pathogen strains.

Project description:Weeds surrounding crops may act as alternative hosts, playing important epidemiological roles as virus reservoirs and impacting virus evolution. We used high-throughput sequencing to identify viruses in Spanish melon crops and plants belonging to three pluriannual weed species, Ecballium elaterium, Malva sylvestris, and Solanum nigrum, sampled at the edges of the crops. Melon and E. elaterium, both belonging to the family Cucurbitaceae, shared three virus species, whereas there was no virus species overlap between melon and the other two weeds. The diversity of cucurbit aphid-borne yellows virus (CABYV) and tomato leaf curl New Delhi virus (ToLCNDV), both in melon and E. elaterium, was further studied by amplicon sequencing. Phylogenetic and population genetics analyses showed that the CABYV population was structured by the host, identifying three sites in the CABYV RNA-dependent RNA polymerase under positive selection, perhaps reflecting host adaptation. The ToLCNDV population was much less diverse than the CABYV one, likely as a consequence of the relatively recent introduction of ToLCNDV in Spain. In spite of its low diversity, we identified geographical but no host differentiation for ToLCNDV. Potential virus migration fluxes between E. elaterium and melon plants were also analyzed. For CABYV, no evidence of migration between the populations of the two hosts was found, whereas important fluxes were identified between geographically distant subpopulations for each host. For ToLCNDV, in contrast, evidence of migration from melon to E. elaterium was found, but not the other way around. IMPORTANCE It has been reported that about half of the emerging diseases affecting plants are caused by viruses. Alternative hosts often play critical roles in virus emergence as virus reservoirs, bridging host species that are otherwise unconnected and/or favoring virus diversification. In spite of this, the viromes of potential alternative hosts remain largely unexplored. In the case of crops, pluriannual weeds at the crop edges may play these roles. Here, we took advantage of the power of high-throughput sequencing to characterize the viromes of three weed species frequently found at the edges of melon crops. We identified three viruses shared by melon and the cucurbit weed, with two of them being epidemiologically relevant for melon crops. Further genetic analyses showed that these two viruses had contrasting patterns of diversification and migration, providing an interesting example on the role that weeds may play in the ecology and evolution of viruses affecting crops.

Dataset Information

A novel method for detecting morphologically similar crops and weeds based on the combination of contour masks and filtered Local Binary Pattern operators.

Publications

A novel method for detecting morphologically similar crops and weeds based on the combination of contour masks and filtered Local Binary Pattern operators.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets