MétaCan
Menu
Back to cohort
Record W2770155217 · doi:10.1080/10106049.2017.1408704

Step-wise Land-class Elimination Approach for extracting mixed-type built-up areas of Kolkata megacity

2017· article· en· W2770155217 on OpenAlexaff
Ansar Khan, Soumendu Chatterjee, Hashem Akbari, Saad Saleem Bhatti, Apurba Dinda, Chandana Mitra, Haoyuan Hong, Quang Doan

Bibliographic record

VenueGeocarto International · 2017
Typearticle
Languageen
FieldEnvironmental Science
TopicLand Use and Ecosystem Services
Canadian institutionsConcordia University
FundersU.S. Geological Survey
KeywordsNormalized Difference Vegetation IndexPixelLand coverRemote sensingVegetation (pathology)GeographyLand useComputer scienceEnvironmental scienceArtificial intelligenceClimate changeGeologyEngineering

Abstract

fetched live from OpenAlex

The extraction of urban built-up areas is an important aspect of urban planning and understanding the complex drivers and biophysical mechanism of urban climate processes. However, built-up area extraction using Landsat data is a challenging task due to spatio-temporal dynamics and spatially intermixed nature of Land Use and Land Cover (LULC) in the cities of the developing countries, particularly in tropics. In the light of advantages and drawbacks of the Normalized Difference Built-up Index (NDBI) and Built-up Area Extraction Method (BAEM), a new and simple method i.e. Step-wise Land-class Elimination Approach (SLEA) is proposed for rapid and accurate mapping of urban built-up areas without depending exclusively on the band specific normalized indices, in order to pursue a more generalized approach. It combines the use of a single band layer, Normalized Difference Vegetation Index (NDVI) image and another binary image obtained through Logit model. Based on the spectral designation of the satellite image in use, a particular band is chosen for identification of water pixels. The Double-window Flexible Pace Search (DFPS) approach is employed for finding the optimum threshold value that segments the selected band image into water and non-water categories. The water pixels are then eliminated from the original image. The vegetation pixels are similarly identified using the NDVI image and eliminated. The residual pixels left after elimination of water and vegetation categories belong either to the built-up areas or to bare land categories. Logit model is used for separation of the built-up areas from bare lands. The effectiveness of this method was tested through the mapping of built-up areas of the Kolkata Metropolitan Area (KMA), India from Thematic Mapper (TM) images of 2000, 2005 and 2010, and Operational Land Imager (OLI) image of 2015. Results of the proposed SLEA were 95.33% accurate on the whole, while those derived by the NDBI and BAEM approaches returned an overall accuracy of 83.67% and 89.33%, respectively. Comparisons of the results obtained using this method with those obtained from NDBI and BAEM approaches demonstrate that the proposed approach is quite reliable. The SLEA generates new patterns of evidence and hypotheses for built-up areas extraction research, providing an integral link with statistical science and encouraging trans-disciplinary collaborations to build robust knowledge and problem solving capacity in urban areas. It also brings landscape architecture, urban and regional planning, landscape and ecological engineering, and other practice-oriented fields to bear together in processes for identifying problems and analyzing, synthesizng, and evaluating desirable alternatives for urban change. This method produced very accurate results in a more efficient manner compared to the earlier built-up area extraction approaches for the landscape and urban planning.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.034
Threshold uncertainty score0.068

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0030.001
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0020.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.031
GPT teacher head0.275
Teacher spread0.244 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations26
Published2017
Admission routes1
Has abstractyes

Explore more

Same venueGeocarto InternationalSame topicLand Use and Ecosystem ServicesFrench-language works237,207