Comparison of Pixel and Object Oriented Based Classification of Hyperspectral Pansharpened Images

Zoleikani, R.; Zoej, Mohammad Javad Valadan; Mokhtarzadeh, M.

doi:10.1007/s12524-016-0573-6

Cited by 13 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The default settings for the “composition of homogeneity criterion” were: color factor of 0.9, shape factor of 0.1, compactness factor of 0.2, and smoothness factor of 0.8. These parameters were selected using a trial and error empirical analysis (Moeller and Stefanoy, 2004; Zoleikani et al, 2017), which choose the most suitable parameters to detect small remnants of vegetation units such as the mangroves (fine scale) and discriminate spread and irregular vegetation units with different seasonal characteristic (high color and smoothness factors).…”

Section: Methodsmentioning

confidence: 99%

A methodology for mapping native and invasive vegetation coverage in archipelagos

Rivas‐Torres

Benítez

Rueda

et al. 2018

Progress in Physical Geography: Earth and Environment

View full text Add to dashboard Cite

This study develops a mixed, systematic, low-cost methodology to define and map native vegetation and the spread of the most aggressive invasive species in islands biomes, focusing on the Galápagos National Park (GNP). Based on preliminary legends defined by experts, Landsat 8/OLI fusion imagery was used for object-oriented classification to obtain the vegetation map of this archipelago. This technique was later verified and validated using high-resolution images from unmanned aerial vehicles (UAVs, i.e., drones) and dedicated satellites, ground truthing, and visual confirmation around GNP coasts. This mixed methodology allowed mapping of nine native ecosystems, six invasive-dominated vegetation units, and two types of lavas. Around 53.63% of GNP is covered by native ecosystems and ∼2.2% is “canopy” dominated by invasive species to date. Native-dominated deciduous forest types cover ∼40.8% of the GNP and only ∼12.8% of the protected area is nowadays covered by humid and transitional type native ecosystems. Among humid native ecosystems, those distributed in the highlands only cover 4.8% and are highly threatened by invasive species, which are mostly distributed in these summit areas. Of the five islands (out of 18) recording invasive-dominated units, Isabela and Santa Cruz were the most infested. Cedrela odorata, Pennisetum purpureum, and Psidium guajava were the main invasive plants dominating the GNP canopy. Highly noxious Rubus niveus was the only invasive species dominating areas among the five infested islands. Methodology detailed here proved useful to provide accurate spatially-explicit islands vegetation data, potential for replication in time, and is expected to aid suitable management of highly endangered and unique biotas in this and other tropical island biomes.

show abstract

Section: Methodsmentioning

confidence: 99%

A methodology for mapping native and invasive vegetation coverage in archipelagos

Rivas‐Torres

Benítez

Rueda

et al. 2018

Progress in Physical Geography: Earth and Environment

View full text Add to dashboard Cite

show abstract

“…The method outperformed the wavelet-based multiresolution analysis method of segmentation with an overall classification accuracy of 91.4–95.4%. Object-based classification Zoleikani et al ( 2017 ) did correlation analysis of pixel and object-based classification of hyperspectral pan-sharpened images (Hyperion and IRS-1C PAN images) for Tehran, Iran, and found that object-oriented procedure presented more accurate outcomes (90.47%) than those obtained by pixel-based classification method (77.33%). It was also observed that Haar wavelet approach has good quality in preserving spectral information as well as spatial information.…”

Section: Discussionmentioning

confidence: 99%

“…• Fuzzy classification Shalan et al (2003) (Pesaresi and Benediktsson 2000). Benediktsson et al (2003) Park et al (1999Park et al ( , 2001, Sanjeevi et al (2001), Saraf (1999), Teggi (2003), Zhang et al (2002) Urban and regional information extraction Ansari and Buddhiraju (2019), Benediktsson et al (2003), Kontoes et al (2000), Pathak and Dikshit (2005), Pesaresi (1999), Pesaresi and Benediktsson (2000), Rashed et al (2001Rashed et al ( , 2005, Shalan et al (2003), Zoleikani et al (2017) Change detection/change monitoring Bochenek (2005), Jain and Sharma (2019), Jensen and Im (2007), Turker and Asik (2005) Topographical mapping Jayaprasad et al (2002), Kumar et al (2004), Rao et al (2003), Srivastava et al (1996), Topan et al (2009), Zhang et al (2002) Scale and mapping issues, potential applications Armenakis and Savopol (1998), Cheng and Toutin (1998), Diwakar et al (2013), Elghazali (2006), Gautam (1997), Jayaraman et al (2000), Jensen and Cowen (1999), Joseph (1996), Meinel et al (1998), Peinsipp-Byma and Roller (1999),…”

Section: Urban and Regional Information Extractionmentioning

confidence: 99%

A Review of IRS-1C Applications in Urban and Regional Studies, and Infrastructure Planning

Kumar

Rao

Ravindranath

et al. 2021

J Indian Soc Remote Sens

View full text Add to dashboard Cite

The multifaceted data from Indian remote sensing satellite (IRS)-1C with Panchromatic (PAN, 5.8 m, 0.50-0.75 lm), Linear Imaging Self-Scanning Sensor-III (LISS-III, 23.5 m, four multi-spectral bands) and Wide Field Sensor (WiFS, 188 m, red and near-infrared bands) sensors onboard, along with stereo imaging and 5-day revisit capability have been effectively utilised for mapping, monitoring, planning and development of urban and regional areas. It made possible to explore urban and regional areas either independently or in combination with other remote sensing data for base map preparation, land use survey and planning, growth modelling, environment and hazards analysis, utilities and infrastructure planning, etc. This review article articulates the scale and mapping potential of IRS-1C data for urban and regional areas, data fusion methods and information retrieval based on visual or digital image processing techniques, including advanced classifiers. It recapitulates the application potential demonstrated in last 25 years in urban and regional studies, and for infrastructure planning.

show abstract

“…• Per-Pixel Classifiers: By considering the spectral similarities of a pixel with classes 73,107 it will be assigned to a class based on either parametric or non-parametric. Neural networks, SVM, and decision trees are examples of suitable techniques to enhance classifications 59 based on the per-pixel method.…”

Section: Classification Approachesmentioning

confidence: 99%

“…• Object-Oriented Classifiers: The object-based classifier not only considers the spectral values stored in digital number (DN) but also counts on topologic (as neighborhood, contextual) and geometric (as size, shape) as classification parameters 16,73,107 . E-Cognition is an example of this type classifier.…”

Section: Classification Approachesmentioning

confidence: 99%

An Intelligent Hybrid Model for Identity Document Classification

Khandan

2021

Preprint

View full text Add to dashboard Cite

Digitization, i.e., the process of converting information into a digital format, may provide various opportunities (e.g., increase in productivity, disaster recovery, and environmentally friendly solutions) and challenges for businesses. In this context, one of the main challenges would be to accurately classify numerous scanned documents uploaded every day by customers as usual business processes. For example, processes in banking (e.g., applying for loans) or the Government Registry of BDM (Births, Deaths, and Marriages) applications may involve uploading several documents such as a driver's license and passport. There are not many studies available to address the challenge as an application of image classification. Although some studies are available which used various methods, a more accurate model is still required. The current study has proposed a robust fusion model to define the type of identity documents accurately. The proposed approach is based on two different methods in which images are classified based on their visual features and text features. A novel model based on statistics and regression has been proposed to calculate the confidence level for the feature-based classifier. A fuzzy-mean fusion model has been proposed to combine the classifier results based on their confidence score. The proposed approach has been implemented using Python and experimentally validated on synthetic and real-world datasets. The performance of the proposed model is evaluated using the Receiver Operating Characteristic (ROC) curve analysis.

show abstract

Comparison of Pixel and Object Oriented Based Classification of Hyperspectral Pansharpened Images

Cited by 13 publications

References 10 publications

A methodology for mapping native and invasive vegetation coverage in archipelagos

A methodology for mapping native and invasive vegetation coverage in archipelagos

A Review of IRS-1C Applications in Urban and Regional Studies, and Infrastructure Planning

An Intelligent Hybrid Model for Identity Document Classification

Contact Info

Product

Resources

About