Classification of paddy crop and weeds using semantic segmentation

Kamath, Radhika; Balachandra, Mamatha; Vardhan, A Harsha; Maheshwari, Ujjwal

doi:10.1080/23311916.2021.2018791

Cited by 30 publications

(13 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over 40 published studies have made use of diverse computer vision methods in GWAS of plants, including Arabidopsis, maize, wheat, rice, sorghum, soybean, and barley, among others ( Xiao et al 2021 ; Affortit et al 2022 ; Guo et al 2022 ). Developmental traits related to biomass and growth of whole plants or specific tissues have commonly been studied using various methods for segmentation of RGB images, including those involving color thresholds ( Das et al 2015 ; Yang et al 2015 ; Al-Tamimi et al 2016 ; Gage et al 2018 ; Guo et al 2018 ; Pham et al 2019 ; Campbell et al 2020 ; Seethepalli et al 2020 ; Ogawa et al 2021 ; Affortit et al 2022 ), machine learning approaches such as support vector machines and random forests ( She et al 2019 ; Zou et al 2019 ; Carlier et al 2022 ; Xu et al 2022 ), and deep convolutional neural networks (DCNNs; Ma et al 2019 ; Yasrab et al 2019 ; Kamath et al 2022 ; Xu et al 2022 ). In recent years, DCNNs similar to that employed in the present study have demonstrated abilities to outperform earlier methods ( Adams et al 2020 ) and have become the dominant approach used for diverse computer vision tasks.…”

Section: Introductionmentioning

confidence: 99%

GWAS supported by computer vision identifies large numbers of candidate regulators of in planta regeneration in Populus trichocarpa

Nagle,

Yuan,

Kaur

et al. 2024

G3: Genes, Genomes, Genetics

View full text Add to dashboard Cite

Plant regeneration is an important dimension of plant propagation and a key step in the production of transgenic plants. However, regeneration capacity varies widely among genotypes and species, the molecular basis of which is largely unknown. While association mapping methods such as genome-wide association studies (GWAS) have long demonstrated abilities to help uncover the genetic basis of trait variation in plants, the abilities of these methods depend on the accuracy and scale of phenotyping. To enable a largescale GWAS of in planta regeneration in the model tree Populus, we developed a phenomics workflow involving semantic segmentation to quantify regenerating plant tissues (callus and shoot) over time. We found the resulting statistics are of highly non-normal distributions, and employed transformations or permutations to avoid violating assumptions of linear models used in GWAS. We report over 200 statistically supported quantitative trait loci (QTLs), with genes encompassing or near to top QTLs including regulators of cell adhesion, stress signaling, and hormone signaling pathways, as well as other diverse functions. Our results encourage models of hormonal signaling during plant regeneration to consider keystone roles of stress-related signaling (e.g., involving jasmonates and salicylic acid) in addition to the auxin and cytokinin pathways commonly considered. The putative regulatory genes and biological processes we identified provide new insights into the biological complexity of plant regeneration, and may serve as new reagents for improving regeneration and transformation of recalcitrant genotypes and species.

show abstract

Section: Introductionmentioning

confidence: 99%

GWAS supported by computer vision identifies large numbers of candidate regulators of in planta regeneration in Populus trichocarpa

Nagle,

Yuan,

Kaur

et al. 2024

G3: Genes, Genomes, Genetics

View full text Add to dashboard Cite

show abstract

“…Lottes et al [35] achieved the semantic segmentation of sugar beet and weeds using FCN with sequence information; Ma et al [36] proposed a SegNet semantic segmentation model based on FCN and achieved high classification accuracy in the segmentation of rice seedlings and weeds. Kamath et al [37] studied semantic segmentation models, such as PSPNet and SegNet, for the recognition of rice crops and weeds, and all obtained good results with over 90% accuracy.…”

Section: Introductionmentioning

confidence: 99%

An Improved U-Net Model Based on Multi-Scale Input and Attention Mechanism: Application for Recognition of Chinese Cabbage and Weed

Wang

Yao

et al. 2023

Sustainability

View full text Add to dashboard Cite

The accurate spraying of herbicides and intelligent mechanical weeding operations are the main ways to reduce the use of chemical pesticides in fields and achieve sustainable agricultural development, and an important prerequisite for achieving these is to identify field crops and weeds accurately and quickly. To this end, a semantic segmentation model based on an improved U-Net is proposed in this paper to address the issue of efficient and accurate identification of vegetable crops and weeds. First, the simplified visual group geometry 16 (VGG16) network is used as the coding network of the improved model, and then, the input images are continuously and naturally down-sampled using the average pooling layer to create feature maps of various sizes, and these feature maps are laterally integrated from the network into the coding network of the improved model. Then, the number of convolutional layers of the decoding network of the model is cut and the efficient channel attention (ECA) is introduced before the feature fusion of the decoding network, so that the feature maps from the jump connection in the encoding network and the up-sampled feature maps in the decoding network pass through the ECA module together before feature fusion. Finally, the study uses the obtained Chinese cabbage and weed images as a dataset to compare the improved model with the original U-Net model and the current commonly used semantic segmentation models PSPNet and DeepLab V3+. The results show that the mean intersection over union and mean pixel accuracy of the improved model increased in comparison to the original U-Net model by 1.41 and 0.72 percentage points, respectively, to 88.96% and 93.05%, and the processing time of a single image increased by 9.36 percentage points to 64.85 ms. In addition, the improved model in this paper has a more accurate segmentation effect on weeds that are close to and overlap with crops compared to the other three comparison models, which is a necessary condition for accurate spraying and accurate weeding. As a result, the improved model in this paper can offer strong technical support for the development of intelligent spraying robots and intelligent weeding robots.

show abstract

“…but could harm maize and wheat seedlings (for example, removing weeds from a field when planting closely). [6][7] An important area of artificial intelligence is deep learning. To distinguish between field weeds and maize, we employ deep learning.…”

Section: Introductionmentioning

confidence: 99%

Weed Identification and Management in Maize: A Smart System for Precision Weed Control

2023

IRJMETS

View full text Add to dashboard Cite

Consumer acceptance of chemical plant protection is declining because to rising public concern over food security and stiffer international regulations for the use of herbicides in the agri-food chain. Applying a treatment just to the weed in Maize patches will provide site-specific weed management. Identification of Maize plants and weeds is a requirement for several parts of precision farming, including on-the-spot herbicide spraying, robotic weeding, and precision mechanical weed management. Although many various approaches have been put out in recent years, this issue still requires more development in terms of the resilience, speed, and accuracy of the algorithms and the identification systems. In the recent years. In the current study, convolutional neural networks (CNNs) were trained using images captured by an RGB camera of Zea mays, Helianthus annuus, Solanum tuberosum, Alopecurus myosuroides, Amaranthus retroflexus, Avena fatua, Chenopodium album, Lamium purpureum, Matricaria chamomila, Setaria spp., Solanum nigrum, and Stellaria media. On a pool of 3000 photos, three different, ResNet-50, and Xception-were modified and trained. The training images were made up of single-species photos of plant stuff. On the testing of the plant detection and weed species classification algorithms, a Top-1 accuracy between 77% and 98% was obtained.The project is succinctly summarized in the abstract, which also gives a quick rundown of its main components. It captures the key elements of the project, such as the problem statement, the objectives, the approach, and the anticipated results. In the example of the project Weed Identification in Maize, the abstract would emphasize the objective of creating a smart system for precision weed control in maize crops. It would discuss the use of image processing and machine learning methods for precisely identifying and managing weeds. The goals of enhancing herbicide use, minimizing the use of herbicides, and improving maize production and quality would also be mentioned in the abstract. The importance of the initiative in improving sustainable agricultural methods will be emphasized as it came to a close.

show abstract

Classification of paddy crop and weeds using semantic segmentation

Cited by 30 publications

References 29 publications

GWAS supported by computer vision identifies large numbers of candidate regulators of in planta regeneration in Populus trichocarpa

GWAS supported by computer vision identifies large numbers of candidate regulators of in planta regeneration in Populus trichocarpa

An Improved U-Net Model Based on Multi-Scale Input and Attention Mechanism: Application for Recognition of Chinese Cabbage and Weed

Weed Identification and Management in Maize: A Smart System for Precision Weed Control

Contact Info

Product

Resources

About