2021
DOI: 10.1016/j.biosystemseng.2021.06.020
|View full text |Cite
|
Sign up to set email alerts
|

Diagnosing the symptoms of sheath blight disease on rice stalk with an in-situ hyperspectral imaging technique

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
10

Relationship

2
8

Authors

Journals

citations
Cited by 21 publications
(8 citation statements)
references
References 19 publications
0
5
0
Order By: Relevance
“…Spectral information covers wavelengths from 450 to 950 nm and is characterized by a high dimension of redundancy between adjacent wavelengths. Excessive redundant spectral information brings great challenges to detection methods and computational complexity ( Hennessy et al, 2020 ; Zhang et al, 2021b ). Therefore, it is necessary to compress the amount of data by a dimensionality reduction method to reduce the cost of subsequent processing on the basis of not dropping the effective feature spectral information ( Poona et al, 2016 ; Sadeghi-Tehran et al, 2021 ).…”
Section: Methodsmentioning
confidence: 99%
“…Spectral information covers wavelengths from 450 to 950 nm and is characterized by a high dimension of redundancy between adjacent wavelengths. Excessive redundant spectral information brings great challenges to detection methods and computational complexity ( Hennessy et al, 2020 ; Zhang et al, 2021b ). Therefore, it is necessary to compress the amount of data by a dimensionality reduction method to reduce the cost of subsequent processing on the basis of not dropping the effective feature spectral information ( Poona et al, 2016 ; Sadeghi-Tehran et al, 2021 ).…”
Section: Methodsmentioning
confidence: 99%
“…The non-plant background removal is important in disease detection with hyperspectral imaging data since it permits the subsequent analysis focusing only on the plant leaves. In detecting rice sheath blight disease with an in-situ hyperspectral imaging data, a more complex strategy was used combining K-means clustering and spectral feature space analysis [29]. While considering the relatively simple experimental environment, the high-quality background removal results can be achieved with this simple method, and the 750 nm band can also be used in subsequent steps.…”
Section: Leaf Abnormal Area Identificationmentioning
confidence: 99%
“…Since plants and background regions in spectral images have significant differences in the near-infrared and red bands, Zhang et al [73] removed the non-plant background using the K-means clustering algorithm before the detection of the scabs of rice sheath blight. Compared to the traditional background removal methods based on pixel threshold segmentation, the clustering-based method can mitigate the issue of irregular edges and the "salt and pepper" phenomenon.…”
Section: Machine Learningmentioning
confidence: 99%