Composite Image with a Geographic Information System Approach

Nugroho, Arief Kelik; Permadi, Ipung

doi:10.1088/1755-1315/406/1/012020

Cited by 2 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The next step is image composite processing to provide a better visual representation. According to 11 , image composite is a combination of various image channels, making it easier to obtain information about a wide variety of objects. Image compositing can be done in image processing software, one of which is ENVI.…”

Section: Image Datamentioning

confidence: 99%

Multispectral data UAV for rice growth phase: a comparison of pixel-based and object-based approach

Sasongko,

Nasrulloh,

Hadi

et al. 2024

Eighth Geoinformation Science Symposium 2023: Geoinformation Science for Sustainable Planet

View full text Add to dashboard Cite

The advancement of remote sensing image acquisition through Unmanned Aerial Vehicles (UAVs) has seen rapid growth in the last five years, particularly in the field of agricultural mapping. The inclusion of multispectral sensors on UAVs holds potential and capabilities for distinguishing different growth stages of rice crops. However, with respect to this objective, there has been limited research investigating pixel-based and object-based classification approaches using multispectral UAV data. This study aims to assess the capabilities of multispectral aerial photos in identifying rice crop growth stages through both pixel-based and object-based classification methods within a portion of the Banyubiru Subdistrict, Semarang Regency. The Support Vector Machine (SVM) method is employed for pixel-based classification, while the object-based classification (OBIA) process employs the Segment Mean Shift algorithm for segmentation. Training samples and data accuracy are obtained through visual interpretation based on the developed orthomosaic data. Four rice crop growth stages are mapped, namely vegetative, reproductive, ripening, and bare-land phases. The two approaches yield differing accuracy performance. The pixel based approach using support vector machine (SVM) achieves an accuracy of 45% with a kappa coefficient of 0.28, whereas the Object Based Image Analysis (OBIA) approach attains an accuracy of 37% with a kappa coefficient of 0.24. The results indicate that, in this case, the pixel-based approach (SVM) demonstrates higher accuracy compared to the Object Based Image Analysis (OBIA) approach. However, the low accuracy indicates the limitations of pixel based image analysis using spectrometer inputs for mapping using UAV datasets.

show abstract

Section: Image Datamentioning

confidence: 99%